Cardiovascular Catheterization and Intervention

Cardiovascular Catheterization and Intervention Mukherjee
A Textbook of Coronary, Peripheral, and Structural Heart Disease Bates

Roffi
Moliterno
Cardiovascular Catheterization and Intervention

Cardiovascular
About the book About the editors
The field of invasive and interventional cardiology is dynamic with frequent Debabrata Mukherjee, M.D.
advances in both technique and technology. An internationally renowned is Professor and Chief of
Catheterization
team of editors and over 100 contributors have shaped this textbook to Cardiovascular Medicine and
provide clinicians with a thorough guide that covers the procedural and Vice-Chairman of Internal
peri-procedural aspects of coronary, peripheral, and structural heart disease Medicine at Texas Tech
diagnostics and interventions. University.
and Intervention
This comprehensive and highly illustrated textbook presents critical Eric R. Bates, M.D. is Professor
information for anyone active in the field of cardiovascular interventions, of Internal Medicine at the
including: University of Michigan.
• Practical suggestions on how to set up a cardiovascular catheterization Marco Roffi, M.D. is

laboratory, choose the right equipment and minimize radiation exposure. Associate Professor of
Medicine and Director of the A Textbook of Coronary, Peripheral,
• A careful analysis of the general principles of percutaneous coronary
and Structural Heart Disease
Interventional Cardiology Unit
interventions, the specific knowledge needed in different clinical scenarios, at the University Hospital of
as well as the patient selection criteria for each invasive procedure. Geneva, Switzerland.
• In-depth coverage of non-coronary interventions, including 13 chapters on
David J. Moliterno, M.D.
peripheral vascular interventions, including carotid artery stenting, as well
is Professor and Vice-
as newer procedures for intracranial stenosis treatment, septal defect repair,
Chair of Internal Medicine
and left atrial appendage closure.
and Physician-in-Chief of
• An incorporation of emerging procedures in structural heart disease, such Cardiovascular Services
as percutaneous aortic valve replacement and mitral valve repair – that at the University of Kentucky.
although not presently mainstream, will likely become an important domain
of interventional cardiologists.
Given the importance of appropriate training and credentialing for clinicians,

the textbook also includes current national guidelines and policies on the
performance of the various procedures.
Edited by
Debabrata Mukherjee
Eric R. Bates
Telephone House, 69-77 Paul Street, London EC2A 4LQ, UK Marco Roffi
52 Vanderbilt Avenue, New York, NY 10017, USA
www.informahealthcare.com
David J. Moliterno
[ram][D:/informa_Publishing/Mukherjee_2400046/z_production/z_3B2_3D_files/978-1-8418-4664-4_CH0000_O.3d]
[14/7/010/18:7:9] [1–18]
Cardiovascular Catheterization
and Intervention
[14/7/010/18:7:9] [1–18]
This page intentionally left blank

[14/7/010/18:7:9] [1–18]
Cardiovascular Catheterization
and Intervention
A Textbook of Coronary, Peripheral,
and Structural Heart Disease
Edited by
Debabrata Mukherjee
Texas Tech University
El Paso, Texas, U.S.A.
Eric R. Bates
University of Michigan
Ann Arbor, Michigan, U.S.A.
Marco Roffi
University Hospital
Geneva, Switzerland
David J. Moliterno
University of Kentucky
Lexington, Kentucky, U.S.A.
[14/7/010/18:7:9] [1–18]
First published in 2010 by Informa Healthcare, Telephone House, 69-77 Paul Street, London EC2A 4LQ, UK.
Simultaneously published in the USA by Informa Healthcare, 52 Vanderbilt Avenue, 7th Floor, New York, NY 10017, USA.
Informa Healthcare is a trading division of Informa UK Ltd. Registered Office: 37–41 Mortimer Street, London W1T 3JH, UK. Registered in
England and Wales number 1072954.
#2010 Informa Healthcare, except as otherwise indicated
No claim to original U.S. Government works
Reprinted material is quoted with permission. Although every effort has been made to ensure that all owners of copyright material have
been acknowledged in this publication, we would be glad to acknowledge in subsequent reprints or editions any omissions brought to our
attention.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any
means, electronic, mechanical, photocopying, recording, or otherwise, unless with the prior written permission of the publisher or in
accordance with the provisions of the Copyright, Designs and Patents Act 1988 or under the terms of any licence permitting limited copying
issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP, UK, or the Copyright Clearance Center, Inc., 222
Rosewood Drive, Danvers, MA 01923, USA (http://www.copyright.com/ or telephone 978-750-8400).
Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without
intent to infringe.
This book contains information from reputable sources and although reasonable efforts have been made to publish accurate information,
the publisher makes no warranties (either express or implied) as to the accuracy or fitness for a particular purpose of the information
or advice contained herein. The publisher wishes to make it clear that any views or opinions expressed in this book by individual authors or
contributors are their personal views and opinions and do not necessarily reflect the views/opinions of the publisher. Any information or
guidance contained in this book is intended for use solely by medical professionals strictly as a supplement to the medical professional’s
own judgement, knowledge of the patient’s medical history, relevant manufacturer’s instructions and the appropriate best practice
guidelines. Because of the rapid advances in medical science, any information or advice on dosages, procedures, or diagnoses should be
independently verified. This book does not indicate whether a particular treatment is appropriate or suitable for a particular individual.
Ultimately it is the sole responsibility of the medical professional to make his or her own professional judgements, so as appropriately to
advise and treat patients. Save for death or personal injury caused by the publisher’s negligence and to the fullest extent otherwise
permitted by law, neither the publisher nor any person engaged or employed by the publisher shall be responsible or liable for any loss,
injury or damage caused to any person or property arising in any way from the use of this book.
A CIP record for this book is available from the British Library.
Library of Congress Cataloging-in-Publication Data available on application
ISBN-13: 978-1-84184-664-4
Orders may be sent to: Informa Healthcare, Sheepen Place, Colchester, Essex CO3 3LP, UK
Telephone: +44 (0)20 7017 5540
Email: CSDhealthcarebooks@informa.com
Website: http://informahealthcarebooks.com/
For corporate sales please contact: CorporateBooksIHC@informa.com

For foreign rights please contact: RightsIHC@informa.com
For reprint permissions please contact: PermissionsIHC@informa.com
Typeset by MPS Limited, A Macmillan Company

Printed and bound in the United Kingdom
[14/7/010/18:7:9] [1–18]
To all the contributors for their hard work, insight, and wisdom in
making this book a great resource for cardiologists, my parents for
their infinite patience, love, and understanding, and who continue
to be my source of inspiration, and to my wonderful wife, Suchandra,
for her love and support.
—Debabrata Mukherjee
I would like to thank my wife Nancy and children Andrew, Lyndsay,

Alexis, and Evan for all their love and support during the writing
of this book.
—Eric R. Bates
To my spouse Muriel and my children Emma, Thomas, and Giulia

with love and gratitude.
—Marco Roffi
To all the terrific faculty and fellows with whom I have worked in the
catheterization laboratory through the years. For sharing their
thoughts, time, and talents, I am grateful.
—David J. Moliterno
[14/7/010/18:7:9] [1–18]

[14/7/010/18:7:9] [1–18]
Preface
The field of interventional cardiovascular medicine continues to rapidly evolve in

both diagnostic and therapeutic arenas. Over the past decade, substantial advances
have been made on many fronts including the development and utilization of
techniques and devices for intravascular and intracardiac imaging, percutaneous
hemodynamic support, drug-eluting stents, embolic protection, mechanical throm-
bectomy, and percutaneous valve repair and replacement. The evolution of newer
drugs and devices challenges the cardiologist to stay abreast of cutting-edge phar-
macological and mechanical strategies for optimal patient care. The Cardiovascular
Catheterization and Intervention: A Textbook of Coronary, Peripheral, and Structural Heart
Disease aims to provide clinicians with comprehensive guidance on the preproce-
dural, procedural, and postprocedural aspects of coronary, peripheral, and structural
heart disease interventions in general situations and in various clinical settings. The
text features evidence-based discussions on patient selection, vascular access, general
principles of interventional cardiology, and postprocedure management of these
patients and serves as a comprehensive, easily accessible reference for busy practi-
tioners and cardiovascular trainees.
Of foremost importance, the topic areas covered are relevant to the daily
practice of interventional cardiology. The book begins with several chapters dedi-
cated to the general concepts associated with cardiac catheterization and interven-
tional cardiovascular medicine. The subsequent chapters focus on hemodynamic
assessment and coronary angiography in general and in specific situations such as
those in pediatric patients and in adults with congenital heart disease. The bulk of the
text addresses coronary and noncoronary interventions including structural heart
disease interventions. Finally, we have included dedicated chapters on credentialing
and organizing prehospital and hospital systems of cardiovascular care. A large
number of high-quality illustrations make this textbook particularly attractive to the
practitioner.
Essential to the quality and appropriateness of the text is the expertise of the
chapter authors. We are fortunate to have assembled a stellar roster of interventional
cardiovascular experts to create this book. The contributing authors from leading
medical centers around the world have collectively performed hundreds of thou-
sands of procedures and published thousands of peer-reviewed manuscripts. We are
greatly indebted to them. The practice of interventional cardiovascular medicine is
exciting, rewarding, and a privilege each of us enjoys. Likewise, it has been our
personal privilege to work with these superb contributors, our colleagues in interven-
tional cardiology, as well as the editorial team at Informa Healthcare. It is our hope
that you will enjoy this book and that it will be a valuable resource to you in
providing the highest quality care to your patients.
Debabrata Mukherjee
Eric R. Bates
Marco Roffi
David J. Moliterno
[14/7/010/18:7:9] [1–18]

[14/7/010/18:7:9] [1–18]
Contents
Preface . . . . . . . . . . . vii
Contributors . . . . . . xiii
1. Introduction to cardiac catheterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Richard A. Lange, Steven R. Bailey, and L. David Hillis
2. Setting up a catheterization laboratory and equipment considerations . . . . . 13

John W. Hirshfeld, Jr.
3. Radiation safety principles in the cardiac catheterization laboratory ....... 19
Thomas M. Bashore
4. Contrast agents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Michael C. Reed and Hitinder S. Gurm
5. Patient selection, preparation, risks, and informed consent . . . . . . . . . . . . . . . . . . 38

David O. Williams and J. Dawn Abbott
6. Conscious sedation (local anesthetics, sedatives, and reversing agents) .... 47
Steven P. Dunn
7. Vasopressors, vasodilators, and antithrombotics
in the catheterization laboratory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Tracy E. Macaulay and David J. Moliterno
8. Vascular access and closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Zoltan G. Turi
9. Right heart catheterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

Franz R. Eberli
10. Pulmonary hypertension—hemodynamic assessment and response
to vasodilators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
Myung H. Park and Vallerie V. McLaughlin
11. Valvular heart disease—measurement of valve orifice
area and quantification of regurgitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Blase A. Carabello
12. Hemodynamic assessment for restriction, constriction, hypertrophic

cardiomyopathy, and cardiac tamponade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Paul Sorajja
13. Endomyocardial biopsy—indications and procedures . . . . . . . . . . . . . . . . . . . . . . . 136

Gregg F. Rosner, Garrick C. Stewart, and Kenneth L. Baughman
14. Pericardiocentesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Carl L. Tommaso
15. Catheterization of the cardiac venous system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

John C. Gurley
[14/7/010/18:7:9] [1–18]
x CONTENTS
16. Coronary arterial anatomy: normal, variants, and well-described

collaterals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
John P. Erwin and Gregory J. Dehmer
17. Diagnostic angiographic catheters: coronary and vascular .................. 194
Ryan Berg and Michael Lim
18. Coronary imaging: angiography, CTA, MRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Joel A. Garcia and John D. Carroll
19. Cardiac catheterization for pediatric patients ................................ 229

Albert P. Rocchini
20. Cardiac catheterization for the adult with complex
congenital heart disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
John F. Rhodes, Jr. and Jorge R. Alegrıá B.
21. Ventriculography and aortography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
José G. Dıéz and James M. Wilson
22. Pulmonary angiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270

Syed Sohail Ali and P. Michael Grossman
23. Transseptal catheterization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
Zoltan G. Turi
24. Mesenteric and renal angiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
Steven J. Filby, Mehdi H. Shishehbor, and Leslie Cho
25. Peripheral arterial angiography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

Ivan P. Casserly
26. Carotid and cerebral angiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
Andrei Pop and Robert D. Safian
27. Application of intracoronary physiology: use of pressure and flow
measurements in clinical practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
Olivier Muller and Bernard De Bruyne
28. Intravascular ultrasound and virtual histology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355

Adriano Caixeta, Akiko Maehara, and Gary S. Mintz
29. Optical coherence tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370

E. Regar, N. Gonzalo, G. Van Soest, and P. W. Serruys
30. Percutaneous coronary intervention—general principles . . . . . . . . . . . . . . . . . . . . 388
Jack P. Chen, Dongming Hou, Lakshmana Pendyala, and Spencer B. King
31. Guiding catheters and wires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404

David W. M. Muller
32. Coronary artery stenting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412

Stéphane Cook and Stephan Windecker
33. Primary PCI in ST elevation myocardial infarction . . . . . . . . . . . . . . . . . . . . . . . . . . 429
Sean P. Javaheri and James E. Tcheng
34. Cardiogenic shock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444

Mark J. Ricciardi
35. A clinical rationale and strategy for chronic total coronary occlusion
revascularization: innovative solutions to an old problem . . . . . . . . . . . . . . . . . . 453
Colin M. Barker and David E. Kandzari
[14/7/010/18:7:9] [1–18]
CONTENTS xi
36. Saphenous vein grafts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469

Claudia P. Hochberg and Joseph P. Carrozza
37. Emboli protection devices, atherectomy, thrombus aspiration devices . . . . . 482
Fernando Cura
38. Complications of PCI: stent loss, coronary perforation

and aortic dissection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 492
Amir-Ali Fassa and Marco Roffi
39. Intra-aortic balloon pump counterpulsation, percutaneous

left ventricular support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 504
Troy Weirick, Wendell Ellis, and Richard W. Smalling
40. Intracardiac echocardiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 515
Dennis W. den Uijl, Laurens F. Tops, Nico R. van de Veire, and Jeroen J. Bax
41. TEE to guide interventional cardiac procedures in the catheterization
laboratory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
Matthias Greutmann, Melitta Mezody, and Eric Horlick
42. Percutaneous therapy for aortic valve disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541

Peter Wenaweser, Lutz Buellesfeld, and Eberhard Grube
43. Percutaneous therapies for mitral valve disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554

Ryan D. Christofferson, Patrick L. Whitlow, E. Murat Tuzcu, and Samir R. Kapadia
44. Hypertrophic cardiomyopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 568
Matthew C. Becker, Gus Theodos, Samir R. Kapadia, and E. Murat Tuzcu
45. Atrial appendage exclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577

David R. Holmes, Jr. and Rebecca Fountain
46. Percutaneous closure of atrial septal defect and patent foramen ovale .... 584
Yves Laurent Bayard, Andreas Wahl, and Bernhard Meier
47. Pediatric and adult congenital cardiac interventions ........................ 593
Sawsan M. Awad, Qi-Ling Cao, and Ziyad M. Hijazi
48. Embolization for fistulas and AVMs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602

Nicholas Kipshidze, Robert Rosen, and Irakli Gogorishvili
49. Lower and upper extremity intervention ..................................... 610

Tahir Mohamed and Mark Robbins
50. Renal and mesenteric artery interventions ................................... 625
John H. Rundback and Robert Lookstein
51. Carotid and vertebral artery interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640

Marco Roffi
52. Intracranial stenosis and acute stroke interventions . . . . . . . . . . . . . . . . . . . . . . . . . 659

Alex Abou-Chebl
53. Training program guidelines, case numbers, and maintenance of
certification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 672
Gregory W. Barsness and Charanjit S. Rihal
54. Organizing prehospital and hospital systems of care for STEMI . . . . . . . . . . . 679
Michael D. Miedema, David M. Larson, and Timothy D. Henry
Index . . . . 685
[14/7/010/18:7:9] [1–18]

[14/7/010/18:7:9] [1–18]
Contributors
J. Dawn Abbott Division of Cardiology, Rhode Island Hospital, Warren Alpert

Medical School, Brown University, Providence, Rhode Island, U.S.A.
Alex Abou-Chebl Department of Neurology, University of Louisville, Louisville,

Kentucky, U.S.A.
Jorge R. Alegrı́a B. Gill Heart Institute and Division of Cardiovascular Medicine,

University of Kentucky, Lexington, Kentucky, U.S.A.
Syed Sohail Ali University of Michigan Cardiovascular Center, Ann Arbor,

Michigan, U.S.A.
Sawsan M. Awad Rush University Medical Center, Chicago, Illinois, U.S.A.
Steven R. Bailey Department of Medicine, Division of Cardiology, University of

Texas Health Science Center, San Antonio, Texas, U.S.A.
Colin M. Barker Division of Cardiology, University of Texas Medical School at

Houston and The Heart and Vascular Institute, Memorial Hermann Hospital at the
Texas Medical Center, Houston, Texas, U.S.A.
Gregory W. Barsness Division of Cardiology, Mayo Clinic, Rochester, Minnesota,

U.S.A.
Thomas M. Bashore Division of Cardiology, Duke University Medical Center,

Durham, North Carolina, U.S.A.
Kenneth L. Baughman{ Division of Cardiovascular Medicine, Brigham and

Women’s Hospital, Boston, Massachusetts, U.S.A.
Jeroen J. Bax Department of Cardiology, Leiden University Medical Center, Leiden,

The Netherlands
Yves Laurent Bayard Cardiology, University Hospital Bern, Bern, Switzerland
Matthew C. Becker Department of Cardiovascular Medicine, Cleveland Clinic,

Cleveland, Ohio, U.S.A.
Ryan Berg Division of Cardiology, Saint Louis University School of Medicine,

St. Louis, Missouri, U.S.A.
Lutz Buellesfeld Cardiology, University Hospital Bern, Bern, Switzerland
Adriano Caixeta Interventional Cardiology Department, Hospital Israelita Albert

Einstein, São Paulo, Brazil
Qi-Ling Cao Division of Cardiology, Rush University Medical Center, Chicago,

Illinois, U.S.A.
Blase A. Carabello Division of Cardiology, Baylor College of Medicine and the

Michael E. DeBakey VA Medical Center, Houston, Texas, U.S.A.
John D. Carroll Department of Medicine, Division of Cardiology, University of

Colorado Hospital, University of Colorado Denver, Denver, Colorado, U.S.A.
{
Deceased.
[14/7/010/18:7:9] [1–18]
xiv CONTRIBUTORS
Joseph P. Carrozza Cardiovascular Medicine–Caritas St. Elizabeth’s Medical

Center, Brighton; Caritas Cardiovascular Network; and Tufts University School of
Medicine, Boston, Massachusetts, U.S.A.
Ivan P. Casserly Division of Cardiology, University of Colorado Hospital, Denver,

Colorado, U.S.A.
Jack P. Chen Saint Joseph’s Heart and Vascular Institute, Saint Joseph’s Hospital of
Atlanta, Atlanta, Georgia, U.S.A.
Leslie Cho Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio, U.S.A.
Ryan D. Christofferson Department of Cardiovascular Medicine, Cleveland Clinic,

Stéphane Cook Cardiology, Cantonal Hospital and University of Fribourg,

Fribourg, Switzerland
Fernando Cura Interventional Cardiology Instituto Cardiovascular de Buenos

Aires, Buenos Aires, Argentina
Gregory J. Dehmer Division of Cardiology, Department of Medicine, Scott & White

Healthcare; Texas A&M Health Science Center, College of Medicine, Temple,
Texas, U.S.A.
Bernard De Bruyne Cardiovascular Center Aalst, O.L.V. Clinic, Aalst, Belgium
José G. Dı́ez St. Luke’s Episcopal Hospital/Texas Heart Institute, Baylor College of
Medicine, Houston, Texas, U.S.A.
Steven P. Dunn Department of Pharmacy, Gill Heart Institute and Division of

Cardiovascular Medicine, University of Kentucky, Lexington, Kentucky, U.S.A.
Franz R. Eberli Division of Cardiology, Triemli Hospital, Zurich, Switzerland
Wendell Ellis Division of Cardiovascular Medicine, University of Texas Medical

School at Houston and the Memorial Hermann Heart and Vascular Institute,
Houston, Texas, U.S.A.
John P. Erwin Division of Cardiology, Department of Medicine, Scott & White

Healthcare; Texas A&M Health Science Center, College of Medicine, Temple, Texas,
U.S.A.
Amir-Ali Fassa Division of Cardiology, University Hospital, Geneva, Switzerland
Steven J. Filby Pinehurst Medical Clinic, Pinehurst, North Carolina, U.S.A.
Rebecca Fountain Division of Cardiovascular Diseases, Mayo Clinic, Rochester,

Minnesota, U.S.A.
Joel A. Garcia Department of Medicine, Division of Cardiology, University of

Colorado Hospital, and Denver Health Medical Center, University of Colorado
Denver, Denver, Colorado, U.S.A.
Irakli Gogorishvili JoAnn Medical Center, Tbilisi, Georgia
N. Gonzalo Department of Interventional Cardiology, Thoraxcenter, Erasmus

Medical Center, Rotterdam, The Netherlands
Matthias Greutmann Department of Cardiology, University Hospital, Zurich,

Switzerland
P. Michael Grossman University of Michigan Cardiovascular Center and VA Ann

Arbor Medical Center, Ann Arbor, Michigan, U.S.A.
[14/7/010/18:7:9] [1–18]
CONTRIBUTORS xv
Eberhard Grube International Heart Center Rhein-Ruhr, Essen, Germany
John C. Gurley Gill Heart Institute and Division of Cardiovascular Medicine,

Hitinder S. Gurm Division of Cardiology, University of Michigan Cardiovascular

Center, Ann Arbor, Michigan, U.S.A.
Timothy D. Henry Minneapolis Heart Institute Foundation at Abbott Northwestern

Hospital and University of Minnesota, Minneapolis, Minnesota, U.S.A.
Ziyad M. Hijazi Division of Pediatric Cardiology, Rush University Medical Center,

Chicago, Illinois, U.S.A.
L. David Hillis Department of Medicine, Division of Cardiology, University of

John W. Hirshfeld, Jr. Department of Medicine, Cardiovascular Medicine Division,

Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.
Claudia P. Hochberg Division of Cardiology, Boston Medical Center, and

Department of Clinical Cardiology, Boston University School of Medicine, Boston,
Massachusetts, U.S.A.
David R. Holmes, Jr. Division of Cardiovascular Diseases, Mayo Clinic, Rochester,

Minnesota, U.S.A.
Eric Horlick Structural Heart Disease Intervention Service, Division of Cardiology,

Peter Munk Cardiac Centre, Toronto General Hospital—University Health Network,
Toronto, Ontario, Canada
Dongming Hou Saint Joseph’s Translational Research Institute, Atlanta, Georgia,

U.S.A.
Sean P. Javaheri Department of Cardiology, Dwight David Eisenhower Army

Medical Center, Fort Gordon, Georgia, U.S.A.
David E. Kandzari Division of Cardiology, Scripps Clinic, La Jolla, California, U.S.A.
Samir R. Kapadia Department of Cardiovascular Medicine, Cleveland Clinic,

Spencer B. King Saint Joseph’s Heart and Vascular Institute, Saint Joseph’s
Hospital of Atlanta, Atlanta, Georgia, U.S.A.
Nicholas Kipshidze Lenox Hill Heart and Vascular Institute of New York, New
York, New York, U.S.A. and Kipshidze Central University Hospital, Tbilisi, Georgia
Richard A. Lange Department of Medicine, Division of Cardiology, University of

David M. Larson Minneapolis Heart Institute Foundation at Abbott Northwestern

Hospital, Minneapolis, Minnesota, U.S.A.
Michael Lim Division of Cardiology, Saint Louis University School of Medicine,

St. Louis, Missouri, U.S.A.
Robert Lookstein Division of Interventional Radiology, Mount Sinai Medical

Center, Mount Sinai School of Medicine, New York, New York, U.S.A.
Tracy E. Macaulay Department of Pharmacy, Gill Heart Institute and Division of

Cardiovascular Medicine, University of Kentucky, Lexington, Kentucky, U.S.A.
Akiko Maehara Cardiovascular Research Foundation, New York, New York,

U.S.A.
[14/7/010/18:7:9] [1–18]
xvi CONTRIBUTORS
Vallerie V. McLaughlin Department of Medicine, Division of Cardiovascular

Medicine, University of Michigan Health System, Ann Arbor, Michigan, U.S.A.
Bernhard Meier Cardiology, University Hospital Bern, Bern, Switzerland
Melitta Mezody Division of Cardiology, Peter Munk Cardiac Centre, Toronto

Western Hospital—University Health Network, Toronto, Ontario, Canada
Michael D. Miedema Department of Cardiovascular Medicine, University of

Minnesota, Minneapolis, Minnesota, U.S.A.
Gary S. Mintz Cardiovascular Research Foundation, New York, New York, U.S.A.
Tahir Mohamed Vanderbilt Heart and Vascular Institute, Nashville, Tennessee,

U.S.A.
David J. Moliterno Gill Heart Institute and Division of Cardiovascular Medicine,

David W. M. Muller St. Vincent’s Hospital, Sydney, and University of New South
Wales, New South Wales, Australia
Olivier Muller Cardiovascular Center Aalst, O.L.V. Clinic, Aalst, Belgium
Myung H. Park Department of Medicine, Division of Cardiology, University of

Maryland School of Medicine, Baltimore, Maryland, U.S.A.
Lakshmana Pendyala Saint Joseph’s Translational Research Institute, Atlanta,

Georgia, U.S.A.
Andrei Pop Interventional Cardiology, Department of Cardiovascular Medicine,

William Beaumont Hospital, Royal Oak, Michigan, U.S.A.
Michael C. Reed Division of Cardiovascular Medicine, University of Michigan,

Ann Arbor, Michigan, U.S.A.
E. Regar Department of Interventional Cardiology, Thoraxcenter, Erasmus Medical

Center, Rotterdam, The Netherlands
John F. Rhodes, Jr. Duke Congenital Heart Program, Duke Heart Center, Duke
University Medical Center, Durham, North Carolina, U.S.A.
Mark J. Ricciardi Interventional Cardiology and Cardiac Catheterization Center,

University of New Mexico Health Sciences Center, Albuquerque, New Mexico, U.S.A.
Charanjit S. Rihal Division of Cardiovascular Diseases, Mayo Clinic, Rochester,

Minnesota, U.S.A.
Mark Robbins Vanderbilt Heart and Vascular Institute, Nashville, Tennessee,

U.S.A.
Albert P. Rocchini Division of Pediatric Cardiology, University of Michigan, Ann

Arbor, Michigan, U.S.A.
Marco Roffi Division of Cardiology, University Hospital, Geneva, Switzerland
Robert Rosen Division of Peripheral and Endovascular Interventions,

Interventional Vascular Oncology and Embolization, Lenox Hill Heart and Vascular
Institute of New York, New York, New York, U.S.A.
Gregg F. Rosner Department of Internal Medicine, Brigham and Women’s

Hospital, Boston, Massachusetts, U.S.A.
John H. Rundback Interventional Institute at Holy Name Medical Center, Teaneck,

New Jersey, U.S.A.
[14/7/010/18:7:9] [1–18]
CONTRIBUTORS xvii
Robert D. Safian Interventional Cardiology and Vascular Medicine, Department of

Cardiovascular Medicine, William Beaumont Hospital, Royal Oak, Michigan, U.S.A.
P. W. Serruys Department of Interventional Cardiology, Thoraxcenter, Erasmus

Mehdi H. Shishehbor Heart and Vascular Institute, Cleveland Clinic, Cleveland

Ohio, U.S.A.
Richard W. Smalling Division of Cardiovascular Medicine, University of Texas

Medical School at Houston and the Memorial Hermann Heart and Vascular Institute,
Paul Sorajja Division of Cardiovascular Diseases and Internal Medicine, Mayo

Clinic, Rochester, Minnesota, U.S.A.
Garrick C. Stewart Division of Cardiovascular Medicine, Brigham and Women’s

Hospital, Boston, Massachusetts, U.S.A.
James E. Tcheng Duke Translational Medicine Institute and Duke Heart Center,
Duke University Medical Center, Durham, North Carolina, U.S.A.
Gus Theodos Department of Cardiovascular Medicine, Cleveland Clinic,

Carl L. Tommaso Cardiac Catheterization Laboratory, Skokie Hospital, NSUHS,

Skokie, Illinois, U.S.A.
Laurens F. Tops Department of Cardiology, Leiden University Medical Center,

Leiden, The Netherlands
Zoltan G. Turi Division of Cardiology, Cooper University Hospital, Robert Wood

Johnson Medical School, Camden, New Jersey, U.S.A.
E. Murat Tuzcu Department of Cardiovascular Medicine, Cleveland Clinic,

Dennis W. den Uijl Department of Cardiology, Leiden University Medical Center,

G. Van Soest Department of Biomedical Engineering, Thoraxcenter, Erasmus

Nico R. van de Veire Department of Cardiology, Leiden University Medical Center,

Andreas Wahl Cardiology, University Hospital Bern, Bern, Switzerland
Troy Weirick Division of Cardiovascular Medicine, University of Texas Medical

School at Houston and the Memorial Hermann Heart and Vascular Institute,
Peter Wenaweser Cardiology, University Hospital Bern, Bern, Switzerland
Patrick L. Whitlow Department of Cardiovascular Medicine, Cleveland Clinic,

David O. Williams Division of Cardiovascular Medicine, Brigham and Women’s

Hospital, Harvard Medical School, Boston, Massachusetts, U.S.A.
James M. Wilson St. Luke’s Episcopal Hospital/Texas Heart Institute, Baylor

College of Medicine, Houston, Texas, U.S.A.
Stephan Windecker Cardiology, University Hospital Bern, Bern, Switzerland

[14/7/010/18:7:9] [1–18]

[2/7/010/7:16:22] [1–12]
Introduction to cardiac catheterization

Richard A. Lange, Steven R. Bailey, and L. David Hillis
INTRODUCTION application of scientific methods to the study of cardiac phys-

Cardiac catheterization is one of the most frequently performed iology demonstrated the potential importance of cardiac cath-
procedures in the United States. Over the past 25 years, the eterization and initiated an era of cardiovascular physiologic
number of procedures has increased 3.5-fold because of investigation.
expanded indications and improvements in techniques and In 1861, Etienne Jules Marey, another French physiolo-
equipment (1). In 2005, an estimated 1,322,000 inpatient left gist, in collaboration with Jean Baptiste Aguste Chauveau, a
heart diagnostic catheterizations and 1,271,000 inpatient percu- veterinarian, elucidated the nature of the apex beat by simulta-
taneous coronary interventional (PCI) procedures were per- neously recording its movement and the right atrial and right
formed in the United States (1). According to the most recent ventricular pressures of a conscious horse (7). Their observation
American Hospital Association survey, 36% of the 4836 hospi- that the apical impulse was caused by early forceful ventricular
tals in the United States have adult diagnostic catheterization contraction remains a milestone as the first graphic recording of
laboratories. Of the 1728 hospital-centered adult diagnostic intracardiac events in a conscious animal (Fig. 1.2) (8). Gaining
laboratories, 78% are PCI capable (2). access to the left ventricle via the carotid artery, they studied
left ventricular pressure waveforms and characterized various
phases of the cardiac cycle, and they were the first to obtain
HISTORICAL PERSPECTIVE simultaneous recordings of left ventricular and aortic pres-
The Early Years of Catheterization: Animal and sures. In addition, Marey and Chaveau invented the double-
Cadaveric Studies (1711–1927) lumen catheter, with which they simultaneously measured
The earliest known cardiac catheterization of an animal was pressures in contiguous cardiac chambers in horses and dogs.
performed in 1710 by Reverend Stephen Hales, an English Decades later, Andre Cournand (see following text) utilized
physiologist and parson, when he “bled a sheep to death and this catheter design to perform similar studies in humans.
then led a gun-barrel from the neck vessels into the still-beating In 1870, Adolph Fick (Fig. 1.3), a German physicist and
heart. Through this, he filled the hollow chambers with molten physiologist, proposed a direct method of measuring cardiac
wax and then measured from the resultant cast the volume of output. In a commentary to his local medical society that
the heartbeat and the minute-volume of the heart, which he resulted in a publication less than one page long (the Fick
calculated from the pulse beat” (3,4). In addition, Hales was the principle, 1870), Fick proposed that the cardiac output could be
first to determine systemic arterial pressure, when, in 1727, measured by dividing the oxygen uptake by the corresponding
he measured the rise in a column of blood in a long glass tube arteriovenous oxygen content difference (9). Interestingly, in
secured in an artery (Fig. 1.1). Brass pipes placed in the carotid 1873 Fick published the results of right and left heart catheter-
artery and jugular vein of a horse were connected to an 11-ft- izations that he performed in animals, but he did not utilize or
high glass tube for pressure measurements, with the trachea of validate his method (10). Only two decades later did physiol-
a goose used as a flexible connector. Hales astutely noted that ogists begin to apply the Fick principle in animals. Although
the pressure was different in arteries and veins (blood from the Grehant and Quinquand published a brief report describing
carotid artery rose to a height of >8 ft in the glass tube, whereas use of the technique in dogs in 1886, it was not until 1898, when
blood from the jugular vein rose <1 ft) and between contrac- Zuntz and Hageman made a detailed study of cardiac output in
tions and relaxations of the heart. the horse at rest and during exercise, that the Fick method was
In 1844, the term cardiac catheterization was coined by established as reliable and reproducible (11). Application of the
Claude Bernard, a French physiologist who inserted long glass Fick principle in humans was hindered by the difficulty of
thermometers into a horse’s right and left ventricles from its obtaining samples of mixed venous blood. Invasive catheter-
jugular vein and carotid artery, respectively. By demonstrating ization was thought to be too dangerous to be applied to
that blood temperature was higher in the right ventricle than in human subjects, because of excessive blood loss and the risk
the left, he established that “chemical reactions” (i.e., metabo- of infection. Furthermore, radiography was not yet available, so
lism) occurred in the body rather than the lungs. Subsequently, attempts to obtain samples of mixed venous blood were made
he used this technique to acquire blood samples from various by such avant-garde interventions as direct transthoracic nee-
arterial and venous sites for metabolic studies, and he per- dle puncture of the right ventricle (12).
formed intracardiac pressure recordings in dogs and sheep to With the discovery of X rays in 1895 by William Roent-
study the regulation of systemic arterial pressure by the ner- gen, radiographic equipment was quickly introduced into
vous system. He was the first to describe right and left heart laboratories in universities and medical schools throughout
catheterization via the femoral vein and artery (5,6). Although Europe and North America (13). Within a month of the publi-
Bernard was not the first to perform catheterization, his careful cation of Roentgen’s paper, two European physicians published
[2/7/010/7:16:22] [1–12]
2 CARDIOVASCULAR CATHETERIZATION AND INTERVENTION
Figure 1.3 Adolph Fick (1829–1901), a mathematician, physicist,

Figure 1.1 First documented cardiac catheterization performed by and physiologist. His major research interest was in the physiology
Hales in 1711. Source: Courtesy of the Bettmann Archives. of muscular contraction. He described the calculation of cardiac
output as an outgrowth of his mathematical approach to physiologic
events.
Figure 1.2 First published records of pressure pulses in a cardiac

chamber obtained by catheterization of the right atrium (O) and
ventricle (V) via the jugular vein in an unanesthetized horse. The
third tracing (C) is from an intrathoracic balloon catheter placed to
detect the cardiac impulse. Source: From Ref. 8.
Figure 1.4 Postmortem roentgenogram of the excised heart of a

20-year-old man who died of pneumonitis. The coronary arteries
the first arteriogram, which was obtained by injecting chalk
were injected with a mixture of red lead and gelatin. Source: From
into the brachial artery of a cadaver and observing the arterial Ref. 16.
supply of the hand with a roentgenogram (14). In October 1896,
Francis Williams, a radiologist at Boston City Hospital, pub-
lished his observations on the use of the fluoroscope to evaluate
the beating heart (15). A decade later, Friedrich Jamin and
Hermann Merkel, German physicians, published the first roent- observations served to advance our understanding of the
genographic atlas of human coronary arteries (16). In their 1907 pathophysiology of coronary artery disease, angina pectoris,
description of 29 excised hearts, coronary arteries were injected and myocardial infarction (MI) (17,18).
with a suspension of red lead in gelatin, after which stereo- For angiography to be safe and practical in a living
scopic roentgenograms were obtained (Fig. 1.4) (13). In some of human subject, several developments were necessary, includ-
the specimens, coronary arterial obstructions and collateral ing the availability of a nontoxic radiopaque material that could
blood vessels were noted, which helped form the basis of be safely injected intravascularly. In this regard, an important
James Herrick’s seminal 1912 paper, titled “Certain Clinical advance occurred in 1921 with the introduction of lipiodol, a
Features of Sudden Obstruction of the Coronary Arteries.” His 40% solution of iodine in poppy oil (19). The injection of
[2/7/010/7:16:22] [1–12]
INTRODUCTION TO CARDIAC CATHETERIZATION 3
the tip of the catheter passed the axilla and entered the
right atrium, respectively. He could not continue to
move the catheter forward into the right ventricle
because the catheter was not long enough.
Forssmann’s successful procedure was performed under inaus-
picious circumstances. He had discussed his ideas about heart
catheterization with the chief of surgery, who forbade him from
performing studies in humans without preliminary research in
experimental animals to demonstrate the procedure’s safety. To
procure the surgical instruments for the vena sectio, Forssmann
assuaged the scrub nurse’s concerns about the procedure’s
safety by agreeing to allow her to be the subject, as she had
requested (24). He had her lie down on the operating table and
then strapped her arms and legs. While distracting her by
applying iodine to her elbow, he anesthetized his own arm,
performed a cutdown of the antecubital vein, and inserted a
ureteral catheter into it. At that point, he released her and
Figure 1.5 Werner Theodor Otto Forssmann (1904–1979). In

1929, interested in direct injections into the heart for resuscitation,
he demonstrated on himself the feasibility of cardiac catheterization.
In 1956, he shared the Nobel Prize for Physiology or Medicine with
Andre Cournand and Dickinson Richards. From 1950 onward, he
practiced as a urologist.
lipiodol into an antecubital vein permitted roentgenographic

visualization of the pulmonary circulation. Sodium iodid, orig-
inally introduced as a urographic contrast agent in 1918, was
first administered intravascularly by Osborne and colleagues at
the Mayo Clinic in 1923 (20). The following year, Barney Brooks,
a vascular surgeon at Washington University, described the intra-
arterial injection of sodium iodid in patients with suspected
peripheral vascular disease (21). The application of vascular
roentgenography to the coronary arteries did not take place in
the 1920s, in large part because the technique for human heart
catheterization had not yet been described. Figure 1.6 The hospital in Eberswalde where Forssmann per-
formed the first catheterization. Source: From Ref. 23.
Cardiac Catheterization Performed in Humans

(1928–1929)
In 1929, Werner Forssmann (Fig. 1.5) first reported the passage
of a catheter into the heart of a living subject: himself (22). As a
medical student, Forssmann learned of the work of Bernard,
Chauveau, and Marey, and he became interested in using
catheterization for the intracardiac administration of drugs for
attempted cardiac resuscitation (4). Believing that catheterization
could be applied as safely in humans as in animals, he performed
the self-catheterization during his surgical residency at the
Auguste-Viktoira Hospital in Eberswalde, Germany (Fig. 1.6).
According to Forssmann’s daughter (23),
he inserted the [ureteral] catheter through a vena sectio
of the left cubital vein and pushed it up to about 65 cm—
the estimated distance to the right heart. He experienced
a sensation of warmth on the wall of the vein when he
moved the catheter and a slight cough, which he attrib-
uted to stimulating the vagus nerve. With the catheter in
his heart, he walked from the operating room downstairs Figure 1.7 First documented human cardiac catheterization. As a
to the X-ray room. He took X rays [Fig. 1.7] while 25-year-old surgical resident, Forssmann passed a ureteral catheter
moving the catheter with the help of a nurse. This nurse via a left basilic vein cutdown into his right atrium and then took this
held a mirror in front of Forssmann so that he could roentgenogram. Source: From Ref. 22.
observe the position of the catheter and take X rays when
[2/7/010/7:16:22] [1–12]
enlisted her help in assisting him down the stairs to the X-ray catheterization was not routinely considered “a safe and sound
room to perform fluoroscopy. When word of Forssmann’s self- procedure to study cardiac physiology” (34), so their article
catheterization reached the chief of surgery, Forssmann was served as a breakthrough.
reprimanded for his disobedience and promptly fired. When his The development of pressure manometers and the double-
work was published the following year (1929), it was acclaimed lumen catheter, through which simultaneous pressures in two
by the popular press but ridiculed and vilified by the medical contiguous heart chambers and large vessels could be recorded,
community (22). He was immediately dismissed from his new served as important advances that allowed Cournand and col-
position at a Berlin hospital and informed that “he could lecture leagues to obtain the first tracings of pressures recorded simulta-
in a circus, but never in a respectable German university” (23). neously from the right ventricle and pulmonary artery (Fig. 1.8)
Before being dismissed, he performed experiments in animals to (35). Their “tracing holds a unique place, since it is the first
demonstrate the value of contrast angiography as a diagnostic demonstration that the tip of a catheter was placed in the pulmo-
tool. He used rabbits in his first experiments and later confessed nary artery of man in order to record pressure pulses” (35).
that “if he had started experimenting with rabbits, he would Cournand and his colleague Dickinson W. Richards developed
never have experimented on himself. When the tip of catheter the technique of heart catheterization for safe and widespread use
touched the rabbit’s endocardium, the electrocardiogram showed “not only to normal man but to patients even in the most severe
temporary cardiac arrest” (23). and acute stages of decompensation” (33). As a result of their
Forssmann ultimately continued his studies in dogs at efforts, cardiopulmonary physiologic investigation accelerated
another institution and reported in 1931 that angiography was rapidly. As noted previously, their seminal work was acknowl-
a safe and useful diagnostic procedure (25). During this time edged when they (along with Werner Forssmann) received the
period, he catheterized himself nine more times in an attempt to 1956 Nobel Prize for Physiology or Medicine (Fig. 1.9).
obtain a publishable angiogram of his heart, but he was unsuc-
cessful. “The response of the academic community ranged from
laughter and disbelief to admiration” (23). He left academic
medicine in 1932 to practice urology, remaining in obscurity
until he, Andre Cournand, and Dickinson Richards were
awarded the 1956 Nobel Prize for Physiology or Medicine, he
for pioneering the procedure and the others for developing its
application. Although Forssmann is credited with being the first
to report heart catheterization in human subjects, Otto Klein, in
fact, should be recognized for performing the first diagnostic
right heart catheterizations. In 1929, Klein performed 11 success-
ful right heart catheterizations, including passage of a catheter
into the right atrium and right ventricle, and he estimated the
cardiac output in his human subjects using the Fick principle (26). Figure 1.8 The first published pressure recordings from the pul-
monary artery of man. The simultaneous tracings in the pulmonary
artery and the right ventricle were obtained with a double-lumen
The Era of Hemodynamic Cardiac Studies (1930–1940s) catheter in a patient with severe pulmonary hypertension (Note: the
Although isolated reports of pulmonary (27,28) and right heart scale is in mmHg). Source: From Ref. 35.
angiography via right atrial injection (29) appeared soon after
Forssmann’s publication, cardiac catheterization progressed
slowly in the 1930s because of the poor quality of the radio-
graphic images and concerns about safety. In a 1932 book on
the measurement of cardiac output, Forssmann’s technique was
considered “not only dangerous to the subject, but useless as
far as cardiac output determinations are concerned . . . . This
method must thus be considered merely a clinical curiosity”
(30). In addition, progress in the field was hindered by the
events leading to World War II that interrupted biomedical
research in Europe. However, in the Western Hemisphere,
Castellano (in Cuba) and Rob and Steinberg (in New York)
obtained high-quality images of all four cardiac chambers and
the great vessels using a rapid intravenous injection of radio-
graphic contrast material (31,32).
As Forssmann noted in his Nobel Prize acceptance
speech in 1956, “A turning point in the history of cardiology
is the year 1941, when Cournand and Ranges made known
their first experiments with the heart catheter as a clinical
method of investigation” (4). They showed that “consistent
values for blood gases could be obtained from the right atrium,
that with this, cardiac output could be reliably and fairly Figure 1.9 Nobel Prize Ceremony, Stockholm, Sweden, Decem-
accurately determined by the Fick principle, and furthermore ber, 1956. Standing at center (left to right) are André Cournand,
that the catheter could be left in place for considerable periods Werner Forssmann, and Dickinson W. Richards.
without harm” (33). Prior to their published studies, cardiac
[2/7/010/7:16:22] [1–12]
While Cournand and Richards used right heart and pul- which might lead to an electrical imbalance and a resultant fatal
monary arterial catheterization primarily to study the pulmonary ventricular arrhythmia. In support of these concerns, Nobel
circulation, Lewis Dexter refined its use in patients with heart Laureate André Cournand reported “his personal experience of
disease. In 1946, he described the placement of a stiff catheter in a 100% fatality rate when contrast was selectively injected in the
the pulmonary capillary “wedge” position, where he obtained coronary arteries of dogs” (45).
blood samples for determination of oxygen saturation; the fol- On October 30, 1958, Mason Sones, a pediatric cardiolo-
lowing year, he reported hemodynamic and oximetric data from gist, inadvertently performed the first selective coronary angio-
normal subjects and those with congenital heart disease (36,37). gram in a 26-year-old patient with aortic regurgitation (46).
The development of new synthetic catheter material in After performing left ventriculography, Sones reported (47):
the 1940s (including polyethylene, nylon, woven Dacron, poly-
urethane, and metal braiding) facilitated heart catheterization, I asked my associate to withdraw the catheter tip across
which was previously performed with modified rubber uro- the aortic valve into the ascending aorta so that we could
logic catheters. In part, the development of such synthetic complete the procedure by performing an aortogram with
material resulted from the difficulty of obtaining rubber during the catheter tip in the ascending aorta. My associate
World War II. Concomitantly, improvements in radiographic complied and we relied on the pressure change from the
equipment and techniques (e.g., rapid series “cut films,” image left ventricle to the ascending aorta without sliding
intensifiers, cineangiography, etc.) paved the way for the sub- the table top back under the 5 inch amplifier to confirm
sequent development of diagnostic and therapeutic left heart the exact location of the tip. I didn’t think this was
catheterization and coronary angiography. Finally, safer and necessary because I was quite certain that the catheter tip
more widely applicable techniques with which to gain access to lay in the ascending aorta just above the aortic valve. My
the aorta were developed during this time. Direct needle access associate, Dr. Royston Lewis, made an injection of 40 cc
of the aorta was abandoned in favor of retrograde catheter- of 90% Hypaque through the catheter. About one second
ization of the femoral, brachial, or radial arteries via cutdown before the injection was initiated, I had the switch to
or percutaneous puncture (6). Although aortography had been initiate a cine run. When the injection began, I was
performed in the late 1940s, the passage of a catheter in a horrified to see the right coronary artery become heavily
retrograde fashion across the aortic valve into the left ventricle opacified and realized the catheter tip was actually inside
was considered to be excessively dangerous. Instead, left the orifice of the dominant right coronary artery. I
ventriculography was performed via direct left ventricular shouted, “Pull it out.” Our combined reaction times to
transthoracic needle puncture (6). accomplish withdrawal of the catheter consumed from 3–
4 seconds which meant that approximately 30 cc of 90%
Hypaque had been delivered into the right coronary
artery. I was of course horrified because I was certain
Left Heart Catheterization and Coronary
the patient would develop ventricular fibrillation. At
Angiography (1950–1960s)
that time we did not have direct current defibrillators
Retrograde left heart catheterization was initially reported in
and knew nothing about the application of closed chest
1950 by Zimmerman, who performed “pull-back” pressure
cardiac massage. I climbed out of the hole and ran
measurements in a patient with aortic regurgitation (38).
around the table looking for a scalpel to open his chest
Although enthusiasm for left heart catheterization escalated,
in order to defibrillate him by direct application of the
the lack of safe and reliable access to the arterial system
paddles of an alternating current defibrillator. I looked at
tempered its use. In 1953, Sven-Ivar Seldinger, a Swedish radi-
the oscilloscope tracing of his electrocardiogram and it
ologist, developed a technique of introducing catheters into the
was evident that he was in asystole rather than in
arterial circulation. “A needle is introduced, a guidewire is
ventricular fibrillation. I knew that an explosive cough
pushed into it, and the needle is removed. The catheter then is
could produce a very effective pressure pulse in the aorta
guided in over the wire, which also is removed” (39,40). This
and hoped that this might push the contrast media
technique, which bears his name (the “Seldinger technique”),
through his myocardial capillary bed. Fortunately, he
continues to be used in virtually all catheterization procedures.
was still conscious and responded to my demand that he
A new era in cardiovascular medicine began in 1958
cough repeatedly. After three to four explosive coughs,
when selective coronary angiography was performed in a
his heart began to beat again with initially a sinus
patient undergoing left heart catheterization. Previously, visu-
bradycardia which accelerated into a sinus tachycardia
alization of the coronary arterial system was obtained non-
within 15 to 20 seconds. He then made a perfectly
selectively by a variety of techniques, including the following:
uneventful recovery with no neurological deficit or
1. Balloon occlusion of the ascending aorta during aortic root other sequelae.
injection of contrast material (41)
The failure of ventricular arrhythmias to materialize dur-
2. Aortic root injection of contrast material enhanced by
ing this inadvertent selective coronary arterial opacification
acetylcholine-induced ventricular arrest (42)
convinced Sones that the human coronary circulation was
3. Phasic injections of contrast material timed to occur during
different from that of dogs, so he continued to perform selective
ventricular diastole (43)
coronary angiography in patients, with access to the arterial
4. Use of various catheters designed to opacify the coronary
system obtained via a brachial artery cutdown. Sones’ seminal
sinuses while directing jets of contrast material toward the
discovery eventually opened the door to coronary artery
coronary arterial ostia (44)
bypass surgery and interventional cardiology. The develop-
Selective coronary angiography was not attempted ment of preformed coronary catheters by Judkins, Amplatz,
because of concerns that it would cause myocardial hypoxia, Schoonmaker, and others (6,48–50); a percutaneous femoral
[2/7/010/7:16:22] [1–12]
arterial approach (51); and sheaths with a hemostatic valve (52)

further refined the technique, which has become standard in
thousands of catheterization laboratories around the world.
Shortly after selective coronary angiography was
adopted, quantitation of left ventricular function and correla-
tion with coronary anatomy transpired. In 1960, Dodge and
coworkers reported methods to determine left ventricular vol-
umes via “angiocardiography” and to quantitate global func-
tion by introducing the concept of ejection fraction (53).
Subsequently, their work provided the basis for the quantita-
tive assessment of regional wall motion.
Concomitant with the development of selective coronary
angiography and quantitative ventriculography, the early foun-
dations for coronary intervention were laid by Charles Dotter, a
radiologist in Portland, Oregon. In 1963, he inadvertently
recanalized an occluded right iliac artery by passing a catheter
through the site of occlusion (54). The following year, Dotter
and his trainee, Melvin Judkins, performed the first intentional
percutaneous transluminal angioplasty on the popliteal artery Figure 1.11 Andreas R. Gruentzig standing in front of his poster
of an 82-year-old woman with gangrene who refused amputa- describing percutaneous dilatation of a canine coronary arterial
tion. The site of severe obstruction was dilated with rigid stenosis at the American Heart Association meeting in Miami,
catheters of increasing diameter, the patient’s leg was salvaged, Florida, in November 1976. Source: From Ref. 55.
and angiography two years later demonstrated a patent vessel
(Fig. 1.10) (54). The “Dotter technique” was not widely
embraced in the United States, since it was considered to be
crude, technically cumbersome, and associated with a high develop a suitable catheter with a nondistensible balloon for
incidence of failure and complications. However, it was widely use in narrowed coronary arteries. He chose polyvinyl chloride
employed in Germany, where Andreas Gruentzig, a young at the suggestion of a professor emeritus of chemistry who was
German angiology fellow, learned the technique in 1969 and working across the street from Gruentzig’s institution, the
conceived of adding a balloon to the Dotter catheter so that it University Hospital of Zurich (55). Using heat molding and
could be applied to other arterial systems. compressed air, he manufactured balloon catheters in his
kitchen at night and on weekends with the help of his wife
and friends (6). After testing these balloon catheters in animals,
The Age of Percutaneous Coronary he performed the first of several hundred successful percuta-
Revascularization (1970–1980s) neous balloon angioplasties of peripheral arteries in human
Attempts to improve on Dotter’s techniques by adding a bal- subjects in February 1974 (56). Subsequently, he refined and
loon to the dilatation catheter were met with limited success, in miniaturized his balloon catheters so that they could be used in
large part because the balloon material was too fragile or too the coronary arterial circulation. Although he conducted exten-
compliant for dilatation of rigid plaques. Gruentzig set out to sive studies in animals and human cadavers (Fig. 1.11), skepti-
cism concerning the utility of coronary balloon angioplasty was
widespread. After successful coronary angioplasty in a dog (57),
the pathologist wrote in his report that Gruentzig should
stop doing whatever he was doing and certainly never
come close to a human being with this terrible balloon
instrument because the canine heart and coronary
arteries looked terrible.
Despite widespread skepticism, Gruentzig performed the
first successful percutaneous transluminal coronary angio-
plasty (PTCA) in a human subject in Zurich, Switzerland, on
September 16, 1977. The patient, a 38-year-old man, had unsta-
ble angina and a discrete stenosis in his proximal left anterior
descending coronary artery. Following two balloon dilatations,
he became free of angina, and when he underwent catheter-
ization at Emory University on September 16, 1987—precisely 10
years after his original procedure—his left anterior descending
coronary artery was widely patent. In 2000, repeat catheterization
performed for chest pain thought to be atypical for angina
Figure 1.10 First percutaneous transluminal angioplasty, performed showed that “the dilated area was pristine”(Fig. 1.12) (55).
by Dotter in 1964, on the left popliteal artery of an 82-year-old woman PTCA was first performed in the United States on March
with gangrene who had refused amputation. Before (left), after (middle), 1, 1978, simultaneously by Richard Myler in San Francisco and
and two years later (right). Source: From Ref. 54. Simon Stertzer in New York (6). It was rapidly adopted by
American cardiologists, but in Europe it was received with
[2/7/010/7:16:22] [1–12]
Figure 1.12 Stenosis before (A), immediately

after (B), 10 years after (C), and 23 years after
(D) the world’s first coronary balloon angioplasty in
a human subject, performed by Andreas Gruentzig
in Zurich, Switzerland, on September 16, 1977. The
functional long-term success is documented by
normal stress tolerance and a normal exercise
ECG (E) of the 61-year-old patient, who underwent
the intervention at the age of 38 years. Source:
From Ref. 55.
skepticism. Gruentzig encountered obstacles that prevented The Stent Era (1990s–2000s)
him from developing the technique in Zurich, so in 1980 he The derivation of the term “stent” is attributed to Dr Charles
moved to Emory University in Atlanta, Georgia, where he Stent (1807–1885) (Fig. 1.13) who developed a material that
continued to refine the procedure until his untimely death in enabled him to secure better dental molds (60). The concept of
1985 in an airplane crash. Several technical developments led to endovascular stents is attributed to Alexis Carrel, 1912 Nobel
the widespread use of PTCA, the most important of which was laureate and vascular surgeon, who implanted glass and metal
the introduction of steerable guidewires, which allowed the tubes into the aorta of dogs (61). Fifty years later, Charles Dotter
application of PTCA to distal coronary arterial stenoses and developed “sleeve” graft devices and metallic coils, which he
tortuous vessels. At the same time, softer, smaller-diameter used in experimental animals, but he never implanted them in
guiding catheters and lower-profile, more flexible balloons patients (6). The recognition that PTCA resulted in acute vessel
were manufactured. Altogether, these technologic advances
resulted in higher rates of success and lower rates of compli-
cations, even though PTCA was being used in patients with
more technically difficult stenoses and in those with multi-
vessel coronary artery disease.
Throughout the 1980s, the balloon catheter remained the
primary device for performing angioplasty, but by the late
1980s, new equipments began to emerge. The first addition,
the directional atherectomy device developed by John Simpson,
was based on the idea that removal of atherosclerotic tissue
would be superior to simple dilation of the artery. Peripheral
arterial atherectomy was first reported in 1985 (58), and the
following year the device was successfully applied to coronary
arteries (59). Although higher procedural success rates were
noted with atherectomy than with balloon angioplasty in
randomized comparisons, atherectomy was not widely
adopted, since it was more tedious and expensive and was
associated with a higher rate of non–Q wave MIs. Subse-
quently, other atherectomy (extraction, rotational, thermal,
laser) and thrombectomy devices were developed. In random- Figure 1.13 Charles Stent, an English dentist, formulated a stable
ized comparisons with balloon angioplasty or stenting (see dental mold. In 1916 a plastic surgeon who utilized the material for
following text), many of the atherectomy devices were associ- facial reconstruction was the first to use the term “stent” when he
ated with higher acute complication rates without improving noted, “The dental composition for this purpose is that put forward
late results. Therefore, their use is currently restricted to coro- by Stent and a mold composed of it is known as a ‘Stent.’” Source:
nary arterial stenoses that are not thought to be amenable to From Ref. 60.
balloon angioplasty or stenting.
[2/7/010/7:16:22] [1–12]
closure in *5% of patients and restenosis in >30% in those in Initially, intensive anticoagulation and antiplatelet ther-
whom it was attempted accelerated the development of an apy were administered to patients in whom intracoronary
endoprosthetic device (e.g., stent) to enhance procedural stents were deployed to prevent acute (in-hospital) and sub-
results, avert vessel closure, and prevent restenosis. The first acute (within 30 days of implantation) stent thrombosis. As a
implantation of coronary arterial stents in humans occurred in result, bleeding complications at the peripheral arterial punc-
1986 in Europe, with insertion of the spring-loaded, self- ture site were common. Two major developments advanced the
expanding Wallstent (6). The following year, a balloon- use of stents. First, the abandonment of warfarin in favor of
mounted, wire coil stent designed by Cesar Gianturco was dual antiplatelet therapy (aspirin and ticlopidine) substantially
implanted at Emory University Hospital in Atlanta, Georgia, reduced the incidence of the aforementioned bleeding compli-
and a slotted tube stent designed by Julio Palmaz was inserted cations. Second, the discovery (with intravascular ultrasound)
in a patient in Sao Paulo, Brazil. The Gianturco-Roubin stent that stents were often not fully expanded and in poor apposi-
was approved for use by the Food and Drug Administration tion against the coronary arterial wall led to the use of high
(FDA) in 1993 in subjects with abrupt and threatened arterial pressure balloon inflations to achieve more complete stent
closure. In 1994, the Palmaz-Schatz stent was approved by the expansion. Both advances improved the acute and chronic
FDA (Fig. 1.14) (62). success rates of stents.
Soon after coronary stents were introduced, it was rec-
ognized that in-stent restenosis from neointimal hyperplasia
occurs in *20% to 25% of patients with discrete, short de novo
stenoses and in as many as 60% of those with small caliber
vessels, long lesions, bifurcation lesions, or diabetes mellitus.
Furthermore, in contradistinction to the discrete restenotic
lesions that sometimes occurred following balloon angioplasty,
in-stent restenosis was noted often to be a much more diffuse
lesion that was not readily amenable to repeat dilatation or
atherectomy. Extensive research was performed in the late
1990s to seek a solution to the problem of in-stent restenosis.
Numerous immunosuppressive and antiplatelet regimens were
evaluated but were unsuccessful in reducing its incidence. In
2001, intracoronary radiation (i.e., brachytherapy) after balloon
dilatation was reported to be moderately successful in treating
in-stent restenosis, but its limitations, including late thrombo-
sis, limited applicability, high cost, and the required presence
of a radiation oncologist during the procedure, rendered it
unsuitable for widespread, routine clinical practice.
In early 2000, stents that eluted antiproliferative pharma-
cologic agents directly into the vessel wall were developed. The
antiproliferative drug was bound to the stent via a polymer
coating, which permitted controlled release of the drug for
days, weeks, or months after stent implantation. A sirolimus-
coated (CYPHER1) stent was approved for clinical use in
Europe in April 2002, and in the United States in May 2003,
following randomized studies, which demonstrated a marked
reduction in restenosis when compared with uncoated (bare
metal) stents (63). A paclitaxel-coated stent (TAXUS1) was
approved for use in Europe in January 2003 and in the United
States in March 2004 following studies that demonstrated that it
too was accompanied by a lower incidence of restenosis in com-
parison with bare metal stents. Subsequently, additional drug-
eluting stents (DES) have been developed, each varying in its
delivery platform, polymer coating, and antiproliferative agent.
At present, DES are utilized for the majority of patients undergo-
ing stent implantation for coronary artery disease because of their
expected very low incidence of in-stent restenosis.
At the same time, devices were developed to improve the
evaluation of arteries, selection of therapies, and success of the
procedures described above. Intravascular ultrasound was
Figure 1.14 (A) Spring-loaded, self-expanding stent (Sigwart developed in the late 1980s but did not achieve widespread
Wallstent, Medinvent SA, Lausanne, Switzerland); (B) prototype of use until the 1990s. Although it had long been known to
the Palmaz stent (Johnson and Johnson Interventional Systems, provide an image of the arterial wall that was not available
Warren, New Jersey, U.S.); (C) balloon-expandable flexible coil with angiography, its use became commonplace with the devel-
stent (Gianturco-Roubin Flex-Stent, Cook, Inc., Bloomington, Indiana, opment of atherectomy devices and intracoronary stenting,
U.S.). Source: From Ref. 62. with which operators needed to evaluate vessel characteristics,
including lumen size, extent and location of plaque, and
[2/7/010/7:16:22] [1–12]
presence and distribution of calcium deposits within the arterial Table 1.1 Complications Associated with Diagnostic Cardiac
wall. For patients with coronary luminal narrowings of indeter- Catheterization
minate hemodynamic significance, methods to assess the physi-
ology of coronary flow were developed, including (i) the Doppler Complication Percentage
flow wire, which can record the velocity of blood flow in the Major
coronary artery and thereby determine if a stenosis is impeding Death <0.1
flow, and (ii) the pressure wire, which can be used to assess the Cerebrovascular accident 0.07
presence or absence of a pressure decline across a coronary Myocardial infarction 0.07
Arrhythmia (life threatening) 0.5
stenosis at rest or after the administration of a vasodilator.
Vascular compromise 0.5–1.5
Anaphylaxis (to contrast material) 0.007
Minor
INDICATIONS AND CONTRAINDICATIONS Hives 2.0–3.0
During its early years, cardiac catheterization was performed Nausea/vomiting *5.0
sparingly and with substantial risk. As time has elapsed, con- Vasovagal reaction 3.0
siderable advances have occurred, and the associated morbid-
ity and mortality have fallen precipitously. Today, diagnostic
cardiac catheterization is performed with minimal risk, and
therapeutic catheterization (i.e., PCI and valvuloplasty) is per-
formed without incident in most patients. Cardiac catheterization contrast material. Individuals with an increased risk of death
now plays a central role in the diagnostic evaluation of the include those with (i) advanced (>70-year-old) or very young
patient with suspected or known cardiac disease, and it offers (<1-year-old) age, (ii) marked functional impairment (class IV
percutaneous therapeutic possibilities in many individuals. angina or heart failure), (iii) severe left ventricular dysfunction
Diagnostic cardiac catheterization is appropriate in sev- or coronary artery disease (particularly left main disease), (iv)
eral circumstances. First, it is indicated to confirm or to exclude severe valvular disease, (v) severe comorbid medical conditions
the presence of a condition already suspected from the history, (i.e., renal, hepatic, or pulmonary disease), or (vi) history of an
physical examination, and/or noninvasive evaluation. In such allergy to radiographic contrast material. Patients with signif-
a circumstance, it allows physicians to both establish the pres- icant narrowing of the left main coronary artery have a sub-
ence and to assess the severity of cardiac disease. Second, stantially greater risk of periprocedural death (2.8%) compared
catheterization is indicated to clarify a confusing or obscure with those without left main stenosis (0.1%) (64).
clinical picture in a patient whose clinical findings and non- Major complications (allergic reaction to radiographic
invasive data are inconclusive. Third, it is performed in some contrast media, cardiogenic shock, cerebrovascular accident,
patients for whom corrective cardiac surgery is contemplated congestive heart failure, cardiac tamponade, and renal failure)
to confirm the suspected abnormality and to exclude associated occurring during or within 24 hours of diagnostic catheter-
abnormalities that might require the surgeon’s attention. ization or PCI occur in 1.3% of patients undergoing a diagnostic
Fourth, catheterization occasionally is performed purely as a study and in 2.3% of those undergoing PCI (1). MI during or
research procedure. immediately following diagnostic catheterization occurs in
Therapeutic catheterization is appropriate in several cir- about 0.07% of patients, but most are small and uncomplicated.
cumstances. Percutaneous coronary revascularization (e.g., Cerebrovascular accidents in the pericatheterization period
angioplasty, rotational atherectomy, or endovascular stenting) may be (i) embolic (from the arterial catheter, guidewire, left
may be indicated in the patient with symptomatic atheroscler- ventricular or atrial thrombus, or dislodged atherosclerotic
otic coronary artery disease whose coronary anatomy is suit- plaque) or (ii) ischemic (i.e., existence of extensive cerebrovas-
able for the procedure. Valvuloplasty is indicated in the subject cular disease that, in association with the hemodynamic alter-
with symptomatic isolated pulmonic stenosis, and it is an ations induced by angiography, leads to inadequate cerebral
acceptable alternative to surgery in the patient with mitral perfusion).
stenosis or aortic stenosis in whom valvular anatomy is suitable Catheterization may result in vascular complications,
and surgery is believed to offer an unfavorable risk-to-benefit such as bleeding at the entry site, retroperitoneal bleeding,
ratio due, for example, to advanced age or comorbid medical vascular access occlusion at the entry site, peripheral emboliza-
conditions (i.e., chronic pulmonary, hepatic, or renal disease or tion, vascular dissection, pseudoaneurysm, and arteriovenous
an underlying malignancy). fistula. The incidence of vascular complications in patients
Catheterization is absolutely contraindicated if a men- undergoing diagnostic catheterization is <0.5%, and in those
tally competent individual does not consent. It is relatively undergoing PCI, it is 2.1% (1,65), with the latter being higher
contraindicated if an intercurrent condition exists that, if cor- because of the use of intensive antiplatelet and anticoagulant
rected, would improve the safety of the procedure. therapy and larger lumen catheters than those used in diag-
nostic catheterization.
Numerous minor complications may cause morbidity but
RISKS AND COMPLICATIONS exert no effect on mortality. Local vascular complications occur
As cardiac catheterization has been more frequently performed, in 0.5% to 2.0% of patients. The incidence is similar for the
the incidence of complications has diminished (Table 1.1). brachial and femoral approaches and somewhat higher for the
However, even in skilled hands, the procedure is not without radial approach. Following arterial catheterization by the bra-
risk. The overall incidence of in-hospital mortality is 0.09% for chial or radial approach, thrombosis, dissection, intimal flap
diagnostic catheterization and 0.8% for PCI (1). Such deaths formation, or subintimal hemorrhage may compromise blood
may be caused by perforation of the heart or great vessels, flow to the hand or arm, and the patient may require throm-
cardiac arrhythmias, acute MI, or anaphylaxis to radiographic bectomy or surgical exploration after catheterization. With the
[2/7/010/7:16:22] [1–12]
percutaneous femoral approach, hemorrhage and/or hema- procedure may be performed at a community hospital without
toma formation at the arterial puncture site are the most com- cardiovascular surgical capability or in a freestanding catheter-
mon problems and, if severe, may require limited surgical ization facility, which may be a fixed structure or a mobile unit.
exploration. Arteriovenous fistulae or pseudoaneurysm forma- Since a freestanding laboratory is not physically attached to a
tion may occur, especially if protracted bleeding at the punc- hospital, quick transportation of a patient by stretcher to a
ture site occurs after sheath removal, as may result from hospital is usually not possible. The most recent Society of
inadequate compression of the femoral vessel, insertion of Cardiac Angiography and Intervention survey of cardiac cath-
large sheaths, severe systemic arterial hypertension, prolonged eterization laboratories (published in 2007) identified 75 non-
heparinization, or administration of thrombolytic or antiplate- hospital-based laboratories in the United States, up from 58 in
let agents. Less commonly, femoral arterial thrombosis occurs, the 1999 survey (71).
which requires immediate thrombectomy. Compression by a Some freestanding catheterization laboratories are pri-
large hematoma or groin clamp may cause local nerve damage. vately owned by physicians. Regardless of who owns and
Local infection may occur at the site of catheter insertion and operates them, all freestanding laboratories should have a
manipulation, but this can usually be treated with meticulous clearly defined working relationship with one or more nearby
wound care and antibiotics. hospitals to facilitate emergency transfer of patients when
The administration of radiographic contrast material may required. Such freestanding facilities must be able to stabilize
cause nausea and vomiting as well as a transient fall in systemic the occasional patient who has a complication, and they must
arterial pressure. Occasionally, such injections are associated have the equipment required for endotracheal intubation and
with allergic reactions of varying severity, and a rare individual ventilator support. The physicians using such facilities should
has anaphylaxis. Interestingly, only 15% of individuals with a be facile in performing endotracheal intubation (since on-site
previous allergic reaction to contrast material have another anesthesiologists are not available) and intra-aortic balloon
adverse reaction with repeat administration, and most of insertion. Quality assurance and quality improvement pro-
these are minor (urticaria, nausea, vomiting) (66). In most grams should be in place and reviewed regularly by an outside
patients with a history of contrast allergy, angiography can be consultant. Currently, only diagnostic procedures (i.e., left or
performed safely; however, premedication with glucocorticos- right heart catheterization, ventriculography, and coronary
teroids and antihistamines as well as use of a different contrast angiography) are performed in freestanding laboratories; per-
agent are usually recommended. The endocardial injection of formance of PCI is restricted by state edict only to hospital-
contrast material during ventriculography (so-called endocardial based laboratories.
staining) may cause ventricular irritability. Finally, use of exces- By providing a setting exclusively for low-risk diagnostic
sive quantities of radiographic contrast material may cause procedures, freestanding facilities can eliminate the long wait-
renal insufficiency, which is usually transient. This is particu- ing periods that sometimes occur with inpatient facilities. In
larly likely to occur in patients with preexisting renal dysfunc- addition, cost savings are touted as one of the advantages of
tion and diabetes mellitus, and its occurrence can be minimized such freestanding facilities (69). In many circumstances, how-
by (i) limiting the amount of contrast material used during ever, the growth of freestanding laboratories has been driven
catheterization on the basis of the patient’s weight and serum by a desire to “capture market share” rather than to improve
creatinine (67), (ii) administering sufficient oral and intrave- patient access to expert catheterization. Since most of these
nous fluids during and after the procedure to insure that facilities are physician owned, the potential exists for financial
the osmotic diuresis caused by the hyperosmolar contrast incentives that inappropriately influence the decision to per-
material does not induce intravascular volume depletion, and form the procedure at such a facility. Other concerns associated
(iii) perhaps by administering nephroprotective agents (i.e., with freestanding facilities include the ability to perform an
N-acetylcysteine) before the procedure (68). adequate caseload to maintain the operators’ skills, limited
experience with recognition and management of complications,
inadequate regulation and quality control, and the time
CARDIAC CATHETERIZATION SETTINGS required to transfer patients to nearby hospitals in the event
Cardiac catheterization procedures were originally only per- of an emergency.
formed on inpatients. Nowadays, however, most elective diag- The authors of the American College of Cardiology/
nostic catheterizations are performed in outpatients in Society for Cardiac Angiography and Interventions (ACC/
laboratories based at a hospital with available cardiovascular SCAI) Clinical Expert Consensus Document on Cardiac Cath-
surgery. These laboratories may be fixed or mobile, with the eterization Laboratory Standards (72) recommended patient
latter being located on the hospital premises. Outpatient catheter- selection criteria for individuals deemed suitable for consider-
ization is widely accepted because of its excellent safety record ation of diagnostic catheterization in an outpatient facility.
when performed in properly selected patients (see following They recommended the following exclusion criteria for adult
text). In 16 studies reporting the results of 20,129 individuals patients:
who underwent diagnostic catheterization at a hospital outpa-
tient laboratory, mortality rates ranged from 0% to 0.3%, MI rates 1. Age >75 years
from 0% to 0.7%, the rate of stroke or transient ischemic attack 2. NYHA class III or IV heart failure
from 0% to 0.4%, the incidence of vascular complications from 0% 3. Acute intermediate- or high-risk ischemic syndromes
to 2%, and the rate of bleeding or hematoma from 0% to 7% (69). 4. Recent MI with postinfarction ischemia
These complication rates are similar to those reported in a very 5. Pulmonary edema thought to be caused by ischemia
large multicenter registry of 222,553 patients who had inpatient 6. Markedly abnormal noninvasive test indicating a high
diagnostic catheterization (70). likelihood of left main or severe multivessel coronary
For patients considered to be at low risk of suffering a artery disease
complication or having extensive coronary artery disease, the 7. Known left main coronary artery disease
[2/7/010/7:16:22] [1–12]
8. Severe valvular dysfunction, especially in the setting of 14. Haschek E LO. Ein beitrag zur praktischen verwerthung der photo-
depressed left ventricular systolic performance graphie nach Roentgen. Wiener Klin Wochenschr 1896; 9:63–64.
9. Patients at increased risk for vascular complications 15. Williams F. A method for more fully determining the outline of
10. Complex adult congenital heart disease the heart by means of the fluorescope together with other uses of
this instrument in medicine. Boston Med Surg J 1896; 135:335.
Patients with any of these exclusion criteria would not be 16. Jamin F, Merkel H. Die Koronararterien des menschlichen Herzens
candidates for catheterization in a freestanding facility, nor unter normalen und pathologischen Verhiiltnissen Jena, 1907.
would any patients considered to be at high risk because of 17. Herrick JB. Certain clinical features of sudden obstruction of the
the presence of comorbid conditions, including the need for coronary arteries. J Am Med Assoc 1912; 59:1757–1762.
18. Herrick JB. Landmark article (JAMA 1912). Clinical features of
anticoagulation therapy, poorly controlled hypertension or dia-
sudden obstruction of the coronary arteries. By James B. Herrick.
betes mellitus, allergy to radiographic contrast material, or JAMA 1983; 250:1757–1765.
renal insufficiency. 19. Sicard JA, Forestier J. Methode radiographique d’exploration de la
Although several case series have reported the rate of cavite epidurale par la lipiodol. Rev Neurol 1921; 37:1264–1266.
complications of diagnostic catheterization in freestanding 20. Osborne ED, Sutherland CG, Scholl AJ Jr., et al. Landmark article
facilities, no randomized trials have compared the complication Feb 10, 1923: roentgenography of urinary tract during excretion of
rates in freestanding facilities with those in hospital-based sodium iodid. By Earl D. Osborne, Charles G. Sutherland, Albert J.
laboratories. The rates of mortality, MI, stroke, and vascular Scholl Jr. and Leonard G. Rowntree. JAMA 1983; 250:2848–2853.
complications in these case series are comparable to those 21. Brooks B. Intra-arterial injection of sodium iodid preliminary
observed in hospital outpatient facilities (69). In recent years, report. J Am Med Assoc 1924; 82:1016–1019.
22. Forssmann W. Sondierung des rechten Herzens. Klin Wochenschr
diagnostic catheterization procedures have been combined
1929; 8:2085–2087.
with PCI if the diagnostic study indicates a need for interven- 23. Forssmann-Falck R. Werner Forssmann: a pioneer of cardiology.
tion. Combining the two procedures may lower the overall cost. Am J Cardiol 1997; 79:651–660.
Patients who undergo diagnostic catheterization at a freestand- 24. King SB III. The development of interventional cardiology. J Am
ing facility or a diagnostic-only hospital do not have the option Coll Cardiol 1998; 31:64B–88B.
of a combined procedure if PCI is deemed appropriate; for 25. Forssmann W. Ueber kontrastdarstellung der hohlen des lebenden
these individuals, the PCI must be performed at a different rechten herzens und der lungenschlagader. Munch Med
institution at a separate time. Approximately 30% of patients Wochenschr 1931; 78:489–492.
who undergo diagnostic catheterization in an outpatient setting 26. Klein O. Zur Bestimmung des zirkulatorischen Minutenvolumens
are referred for subsequent PCI (69). nach dem Fickschen Prinzip. Munchener Medizinische Wochens-
chrift 1930; 77:1311–1312.
27. Moniz E, de Carvalho L, Lima A. La visibilite des vaisseaux
pulmonaires aus rayons X par injection dans l’oreillette droite
REFERENCES de fortes solutions d’Iodure de sodium. Bull Acad Med Paris 1931;
1. Rosamond W, Flegal K, Furie K, et al. Heart disease and stroke 105:627–629.
statistics—2008 update: a report from the American Heart Asso- 28. Perez Ara A. El sondage del corazon derechia su tecnica y
ciation Statistics Committee and Stroke Statistics Subcommittee. aplicationes. Rev Med Cir Habana 1931; 36:491–508.
Circulation 2008; 117:e25–e146. 29. Moniz E, de Carvalho L, Saldanha A. Angiopneumographie. J
2. American Hospital Association. Facilities and services: Health Radiol d’Electrol 1932; 16:469–472.
Forum LLC, 2008. 30. Grollman A. The Cardiac Output of Man in Health and Disease.
3. Burchell HB. Editorial Stephen Hales, September 17,1677–January Baltimore: C. C. Thomas, 1932.
4, 1761. Circulation 1961; 23:1–6. 31. Castellanos A, Pereiras R, Argelio Garcia A. La angio-cardiografia
4. Forssmann W. The role of heart catheterization and angiocardiog- radio-opaca. Arch Estud Clin Habana 1937; 31:523–527.
raphy in the development of modern medicine. In: Nobel Lectures, 32. Robb GP, Steinberg I. A practical method of visualization of the
Physiology or Medicine 1942–1962. Stockholm: Elsevier Publishing chambers of the heart, the pulmonary circulation, and the great
Company, 1956. blood vessels in man. J Clin Invest 1938; 17:507 (abstr).
5. Bernard C, Atlee WF, Robin C. Notes of M. Bernard’s Lectures on 33. Richards DW. The contributions of right heart catheterization to
the Blood; with an Appendix. Philadelphia: Lippincott, Grambo & physiology and medicine, with some observations on the physi-
Co, 1854. opathology of pulmonary heart disease. In: Nobel Lectures, Phys-
6. Mueller RL, Sanborn TA. The history of interventional cardiology: iology or Medicine 1942–1962. Stockholm: Elsevier Publishing
cardiac catheterization, angioplasty, and related interventions. Company, 1956.
Am Heart J 1995; 129:146–172. 34. Cournand A, Ranges HA. Catheterization of the right auricle in
7. Luderitz B. Etienne Jules Marey (1830–1904). J Interv Card Electro- man. Proc Soc Exp Biol Med 1941; 46:462–466.
physiol 2005; 12:91–92. 35. Cournand A. Control of the pulmonary circulation in man with some
8. Classic pages: A. Chaveau and É. J. Marey. Circ Res 1968; 22:96. remarks on methodology. In: Nobel Lectures, Physiology or Medi-
9. Fick A. Uber die Messung des Blutquantums in den Herzventri- cine 1942–1962. Stockholm: Elsevier Publishing Company, 1956.
keln. Phys-Med Ges Wurzburg, 1870. 36. Dexter L, Burwell CS, Haynes FW, et al. Oxygen content of
10. Fick A. Ueber die schwankungen des blutdruckes in verschmde- pulmonary “capillary” blood in unanesthetized human beings. J
nen abschnitten des gefass-systems (zum theil nach versuchen Clin Invest 1946; 25:913 (abstr).
von D Naceff aus Belgrad un Rumanien) Vorhandl der Wurzburg 37. Dexter L, Haynes FW, Burwell CS, et al. Studies of congenital
Physlol Med Gesellsch 1873; 4:223. heart disease: II. the pressure and oxygen content of blood in the
11. Zuntz N, Hagemann O. Untersuchungen iiber den Stoffwechsel right auricle, right ventricle, and pulmonary artery in control
des Pferdes bei Ruhe und Arbeit. Landwirtschaftliche Jahrbucher patients, with observations on the oxygen saturation and source
1898; 27:1–438. of pulmonary ‘capillary’ blood. J Clin Invest 1947; 26:554–560.
12. Fishman AP. A century of pulmonary hemodynamics. Am J 38. Zimmerman HA, Scott RW, Becker NO. Catheterization of the left
Respir Crit Care Med 2004; 170:109–113. side of the heart in man. Circulation 1950; 1:357–359.
13. Fye WB. Coronary arteriography—it took a long time! Circulation 39. Doby T. A tribute to Sven-Ivar Seldinger. AJR Am J Roentgenol
1984; 70:781–787. 1984; 142:1–4.
[2/7/010/7:16:22] [1–12]
40. Seldinger SI. Catheter replacement of the needle in percutaneous 58. Simpson JB, Johnson D, Thapliyal HV, et al. Transluminal athe-
arteriography; a new technique. Acta Radiol 1953; 39:368–376. rectomy: a new approach to the treatment of atherosclerotic vas-
41. Dotter CT, Frische LH. Visualization of the coronary circulation by cular disease. Circulation 1985; 72:146 (abstr).
occlusion aortography: a practical method. Radiology 1958; 71: 59. Simpson JB, Zimmerman J, Selmon MR, et al. Transluminal
502–524. atherectomy: initial clinical results in 27 patients. Circulation
42. Arnulf G. L’arteriographie methodique des arteres coronaires 1986; 74:203 (abstr).
grace a l’utilisation de l’acetylcholine; donnees experimentales et 60. Ring ME. How a dentist’s name became a synonym for a life-saving
cliniques. Bull Acad Natl Med 1958; 142:661–672. device: the story of Dr. Charles Stent. J Hist Dent 2001; 49:77–80.
43. Richards LS, Thal AP. Phasic dye injection control system for 61. Carrel A. Presentation speech. In: Nobel Lecture, Physiology or Med-
coronary arteriography in the human. Surg Gynecol Obstet 1958; icine 1901–1921. Amsterdam: Elsevier Publishing Company, 1967.
107:739–743. 62. Schatz RA. A view of vascular stents. Circulation 1989; 79:445–457.
44. Bellman S, Frank HA, Lambert PB, et al. Coronary arteriography. 63. Morice MC, Serruys PW, Sousa JE, et al. A randomized compar-
I. Differential opacification of the aortic stream by catheters of ison of a sirolimus-eluting stent with a standard stent for coronary
special design–experimental development. N Engl J Med 1960; revascularization. N Engl J Med 2002; 346:1773–1780.
262:325–328. 64. Boehrer JD, Lange RA, Willard JE, et al. Markedly increased
45. Ryan TJ. The coronary angiogram and its seminal contributions to periprocedure mortality of cardiac catheterization in patients
cardiovascular medicine over five decades. Circulation 2002; with severe narrowing of the left main coronary artery. Am J
106:752–756. Cardiol 1992; 70:1388–1390.
46. Sones FMJ. Coronary arteriography. In: Eighth Annual Conven- 65. Applegate RJ, Sacrinty MT, Kutcher MA, et al. Trends in vascular
tion of the American College of Cardiology. Philadelphia, 1959. complications after diagnostic cardiac catheterization and percu-
47. Hurst JW. History of cardiac catheterization. In: King SB III, taneous coronary intervention via the femoral artery, 1988 to 2007.
Douglas JS Jr., eds. Coronary Arteriography and Angioplasty. JACC Cardiovasc Interv 2008; 1:317–326.
New York: McGraw-Hill, 1985:1–9. 66. Brogan WC III, Hillis LD, Lange RA. Contrast agents for cardiac
48. Judkins MP. Selective coronary arteriography. I. A percutaneous catheterization: conceptions and misconceptions. Am Heart J 1991;
transfemoral technic. Radiology 1967; 89:815–824. 122:1129–1135.
49. Schoonmaker FW, King SB III. Coronary arteriography by the 67. Cigarroa RG, Lange RA, Williams RH, et al. Dosing of contrast
single catheter percutaneous femoral technique. Experience in material to prevent contrast nephropathy in patients with renal
6,800 cases. Circulation 1974; 50:735–740. disease. Am J Med 1989; 86:649–652.
50. Wilson WJ, Lee GB, Amplatz K. Biplane selective coronary arte- 68. Marenzi G, Assanelli E, Marana I, et al. N-acetylcysteine and
riography via percutaneous transfemoral approach. Am J Roent- contrast-induced nephropathy in primary angioplasty. N Engl J
genol Radium Ther Nucl Med 1967; 100:332–340. Med 2006; 354:2773–2782.
51. Ricketts HJ, Abrams HL. Percutaneous selective coronary cine 69. Agency for Healthcare Research and Quality (AHRQ). Cardiac
arteriography. JAMA 1962; 181:620–624. Catheterization in Freestanding Clinics: a Review. Rockville, MD:
52. Hillis LD. Percutaneous left heart catheterization and coronary Department of Health and Human Services; 2005:1–120.
arteriography using a femoral artery sheath. Cathet Cardiovasc 70. Johnson LW, Lozner EC, Johnson S, et al. Coronary arteriography
Diagn 1979; 5:393–399. 1984–1987: a report of the Registry of the Society for Cardiac
53. Dodge HT, Sandler H, Ballew DW, et al. The use of biplane Angiography and Interventions. I. Results and complications.
angiocardigraphy for the measurement of left ventricular volume Cathet Cardiovasc Diagn 1989; 17:5–10.
in man. Am Heart J 1960; 60:762–776. 71. Dehmer GJ, Blankenship J, Wharton TP Jr., et al. The current status
54. Dotter CT. Transluminal angioplasty: a long view. Radiology 1980; and future direction of percutaneous coronary intervention with-
135:561–564. out on-site surgical backup: an expert consensus document from
55. King SB 3rd, Meier B. Interventional treatment of coronary heart the Society for Cardiovascular Angiography and Interventions.
disease and peripheral vascular disease. Circulation 2000; 102: Catheter Cardiovasc Interv 2007; 69:471–478.
IV81–IV86. 72. Bashore TM, Bates ER, Kern MJ, et al. American College of
56. Gruentzig A. Perkutane rekanalisation chronischer arterieller vers- Cardiology/Society for Cardiac Angiography and Interventions
chlusse mit einem neuen dilatationskatheter modifikation der clinical expert consensus document on cardiac catheterization
Dotter- Technik. Deutsch Med Wochenschr 1974; 99:2502–2510. laboratory standards: summary of a report of the American Col-
57. Meier B, Gruentzig A. The father of balloon angioplasty—I was lege of Cardiology Task Force on clinical expert consensus docu-
there! In: Beller GA, ed. Cardiosource: American College of ments. Catheter Cardiovasc Interv 2001; 53:281–286.
Cardiology, 2003.
[2/7/010/7:17:52] [13–18]
Setting up a catheterization laboratory

and equipment considerations
John W. Hirshfeld, Jr.
OVERVIEW There are three basic reasons for which these architec-
A cardiac catheterization laboratory is a complex facility that tural relationships are important.
integrates multiple items of equipment and a group of clinical
1. Patients are frequently transported between these facilities.
personnel who have differing skill sets and responsibilities. It is
Many of these patients are critically ill and medically
a component of a hospital that relates to multiple facilities
unstable. Short transportation distances and minimal ele-
within the overall umbrella of the hospital system. In addition,
vator rides are important safety considerations.
the laboratory facility has the potential to be a large hospital
2. Patients who have recently undergone invasive diagnostic
cost center as it utilizes expensive equipment and supplies.
and therapeutic procedures are frequently cared for in
Consequently, it is important that a cardiac catheterization
critical care units. In the event that they develop a problem
laboratory be operated efficiently. This requires careful thought
post procedure, if they are located in close proximity to the
and planning in developing its architectural design, selecting
cardiac catheterization suite, the cardiac catheterization
and integrating its equipment, and developing its operating
laboratory staff can check on them promptly and efficiently.
procedures and protocols.
3. Supporting staff needed to respond to emergencies occur-
ring in the cardiac catheterization suite such as anesthesi-
ARCHITECTURAL CONSIDERATIONS ologists, and respiratory therapists are frequently located
General Considerations in the operating rooms and critical care units. Thus, if these
A cardiac catheterization laboratory suite is an integrated facil- units are in close proximity to the cardiac catheterization
ity that provides care to a variety of patients ranging from laboratory suite, the support personnel will be able to
stable outpatients undergoing diagnostic procedures to criti- respond more promptly.
cally ill patients presenting to the emergency room with ST
elevation myocardial infarctions requiring emergent interven-
tional procedures. This complement of services requires that it Procedure Room Design
be architecturally configured to accommodate the broad range The procedure room is the core of the facility, and its design
of patients. should not be compromised by competing architectural con-
Facilities required in a cardiac catheterization suite include siderations. The first consideration is to provide adequate floor
the following: space. Many state departments of health codes specify minimal
floor areas. In addition, the X-ray equipment manufacturers
1. Procedure rooms with attached control rooms and X-ray
specify minimum room sizes and minimum clearances between
equipment electronics rooms
equipment items and adjacent walls. However, irrespective of
2. Patient pre- and postprocedural care area including patient
whatever code requirements exist, a minimal procedure room
changing areas and lockers
size should be 500 ft2 (more if the room contains a biplane X-ray
3. Equipment and supply storage area
system).
4. Patient registration and family waiting area
In laying out the procedure room, particular attention
5. Data review and report-generating area
should be paid to circulation, sight lines, and relationships with
6. Office space for facility administrative personnel
supporting utilities.
7. Relationship to other hospital facilities
Patient entry locations and circulation space around the
The design of a cardiac catheterization suite should procedure table should be designed to facilitate ease of patient
include development of programmatic requirements that spec- entry and transfer to the procedure table. In planning circula-
ify the number of procedure rooms needed and the space to be tion, one should consider the worst-case scenario of getting a
allocated to each out of procedure room function (1). An addi- patient who is in an oversized hospital bed on mechanical
tional architectural challenge is to arrange the components of ventilation, and circulatory support in and out of the procedure
the suite so that they relate well to each other and provide for room.
efficient circulation between them. Sight line considerations are important to facilitate com-
The cardiac catheterization suite also interacts closely munications and interactions between the procedure room staff
with other hospital facilities. These include critical care units, and the control room staff. Thus, the location of the control
the emergency department, and cardiac operating rooms. Ideal room, its window to the procedure room, and the provision for
architectural relationships place the cardiac catheterization audio communication between the control room and the pro-
suite in close proximity to these facilities. cedure room are an important consideration.
[2/7/010/7:17:52] [13–18]
The X-ray imaging system has four groups of components. must be positioned carefully so that it does not conflict with the
movements of X-ray system components.
1. The X-ray gantry supporting the X-ray tube and imaging
Procedure room lighting is another important consider-
system and the patient procedure table
ation. Considerable attention should be devoted to the type of
2. The X-ray video monitors mounted on a ceiling-suspended
lighting employed and its control system. The laboratory
monitor boom in the procedure room
operates under several different lighting conditions. During
3. The X-ray system’s supporting electronic equipment,
setup, patient entry, and patient preparation, a high lighting
which should be housed in a separate climate-controlled
level is needed. During procedures, dimmer lighting is desir-
room adjacent to the procedure room
able to facilitate viewing the X-ray image monitors. During dim
4. The X-ray control system, which is housed in the control
lighting conditions, it is desirable to have brighter spot lighting
room along with a duplicate set of video monitors
in selected locations such as the catheter entry site area and the
The procedure room, control room, and X-ray electronics instrument table. Ideally, lighting controls that can be operated
equipment room should be designed with adequate space, by the physicians performing the procedure should be fur-
physical relationships, and sight lines to optimize the position- nished at the procedure table. This facilitates selecting the
ing and configuration of the X-ray system. optimal lighting level for a given procedure phase.
The procedure room also needs to be able to accommodate
items of ancillary portable equipment. This includes portable
Procedure Room Utilities imaging equipment (including intravascular and intracardiac
The procedure room has important utility requirements for ultrasound), portable physiologic monitoring equipment
electrical power, oxygen, vacuum, and lighting. (including pressure wire and flow wire consoles), and circula-
The X-ray system is powered by its own dedicated elec- tory support equipment (including intraaortic balloon pump,
trical power source. This is typically 480-V three-phase power Impella and various extracorporeal circulatory support devices).
that is routed separately to the X-ray system. This power is Each of these units requires space when in use and utility
required to operate the X-ray generator, which typically connections. Some require connection to dedicated monitors
requires as much as 150 KW. Most X-ray systems contain located on the X-ray imaging monitor boom. Some require con-
transformers that step electrical power down to lower voltages nection to the laboratory’s physiologic monitoring system to
to power components of the X-ray unit such as the procedure send signals to it or receive signals from it (i.e., ECG signals for
table and the monitors. intraaortic balloon pump and ultrasound machines). Each of
It is important to provide ample receptacles for standard these utility connections and monitor connections should be
110-V single-phase electrical power in both the procedure room specified at the time of procedure room design so that the
and the control room. These receptacles are needed to power appropriate connectors and cabling are designed into the room
portable equipment such as the defibrillator, mechanical ven- at the time of construction.
tilators, ultrasound machines, and other portable items that Since the procedure room is a source of diagnostic
may be used regularly or intermittently in the procedure room. X radiation, its walls must be shielded. State departments of
A basic axiom is that it is impossible to furnish too many health construction codes specify the type and extent of shield-
electrical receptacles as the need for portable devices appears to ing required as well as the X-ray signage requirements.
be inexhaustible. It is also important that receptacles be located
throughout the room on all walls, above counter tops and,
importantly, in the procedure table pedestal. Floor-mounted Control Room Design
receptacles are undesirable because the floor is frequently wet. The design and layout of the control room are among the most
One important consideration in the design of a labora- complex and are most frequently overlooked when designing a
tory’s electrical power system is the provision of uninterrupt- cardiac catheterization laboratory suite. The challenge is to
able or emergency power capability. Superficially, it would arrange all of the equipment needed in the control room (fre-
seem appropriate to have all electrical power supported by the quently provided by different manufacturers) with appropriate
hospital’s emergency power system. However, this approach cabling and ergonomic arrangement to facilitate control room
operations. In addition, the design should provide for future
may require a larger-capacity emergency generator than would
otherwise be required. In addition, consideration must be given changes in equipment configuration. The control room requires a
to the response time for switchover to emergency power. Cur- seating area for the staff monitoring the procedure that has good
rent X-ray and physiologic monitoring systems require long sight lines to the procedure room and to the physiologic mon-
initialization procedures (some as long as 10 minutes). Thus, if itoring and X-ray equipment. There are numerous computer
power is interrupted long enough to require that the system cases and monitors associated with these functions that need to
reinitialize, it is possible that a laboratory X-ray system will be be arranged and cabled appropriately. Cabling includes dedi-
inoperable for as long as 10 minutes after switchover to emer- cated signal cables from the procedure room and the X-ray
gency power. equipment control room as well as hospital network connections
Oxygen and vacuum are essential components of the and electrical power connections. In addition, the room design
laboratory’s emergency response system. The outlets for these needs to provide the flexibility to accommodate future changes
utilities should be positioned close to the head end of the pro- in X-ray imaging or physiologic monitoring equipment.
cedure table to be readily available to apply to a patient. One
location for positioning these utilities is on the wall closest to the Pre- and Postprocedure Care Area Design
patient’s head. There they are typically grouped with some other The pre- and postprocedure care area must provide comfortable
emergency resuscitation equipment. An alternative location is on private areas for individual patients with appropriate staff access
a ceiling-mounted anesthesia column. This latter location offers and monitoring capabilities. The overall purpose and functions
the benefit of being closer to the patient. However, if employed, it provided by these units influence design considerations. At a
[2/7/010/7:17:52] [13–18]
SETTING UP A CATHETERIZATION LABORATORY AND EQUIPMENT CONSIDERATIONS 15
minimum, these units provide a facility for the intake of patients Data from the physiologic monitoring system and report gen-
before procedure and for their short-term aftercare post proce- eration system are similarly transmitted to and from archive
dure. Some units also function as 23-hour stay facilities for servers. The laboratory’s X-ray system and monitoring system
patients undergoing interventional procedures. Thus, the partic- read ADT (admissions, discharge, transfer) demographic and
ulars of design and capabilities provided will vary depending on accounting data from the hospital information system and
the functionalities intended. Anticipated utilization levels will postaccounting data from completed procedures to it. The
determine the total capacity designed into the unit. laboratory also, ideally, makes its reports available throughout
the hospital network for access by physicians and clinical staff
caring for the patient. Both the X-ray system and the physio-
Supply Storage and Inventory Management logic monitoring system require maintenance and troubleshoot-
A cardiac catheterization facility must maintain a large inven-
ing. This is frequently accomplished by vendor service
tory of devices readily accessible on site, as it is not possible to
personnel connecting remotely via network connections to the
anticipate and procure in advance all of the devices that may be
equipment in the procedure rooms.
required to complete a given procedure. Thus, the facility needs
A cardiac catheterization laboratory’s computer network
to have a supply storage area located in close proximity to the
demands, thus, are substantial. These demands must be con-
procedure room(s) so that needed devices can be obtained
sidered in the design of the network backbone of the system.
promptly as the need arises. The overall scale of operation of
The network must have sufficient bandwidth to support real-
the laboratory facility determines the size of this facility.
time communications between the core systems in the proce-
Inventory management is a complex challenge for a car-
dure rooms, the archive servers, and the client terminals used
diac catheterization laboratory. The facility must inventory a
to access images and procedure data outside of the procedure
large range of devices and supplies, some of which are used
rooms. It must also be sufficiently secure to protect against the
regularly with almost every case, while others are used only
intrusion of viruses and other computer malware into the core
sporadically in unusual circumstances. It is important not to
system components. The bandwidth issue is particularly
run out of either commonly or infrequenlty used items because,
important given that hospital information networks experience
in a particular situation, availability of a given item may make
large fluctuations in traffic volume. Consequently, a network
the difference between success and failure in completing a case.
design that is adequate for low traffic periods may choke
One of the challenges of maintaining the inventory of infre-
during high traffic periods. One solution is to isolate the core
quently used items, in addition to replenishing supplies when
network connecting the procedure rooms and the archive
used, is to monitor expiration dates.
servers from the balance of the hospital network so that the
Inventory management begins with establishment of a
LAN performance between procedure rooms and archive serv-
comprehensive list of items to be inventoried and establishment
ers is not affected by traffic elsewhere in the hospital network.
of par levels for each item on the basis of its anticipated usage
Security is an important issue as well. As many X-ray and
pattern and resupply times and reliability. A variety of inven-
physiologic monitoring systems run under a Windows operat-
tory management systems are available to facilitate this pro-
ing environment using open-source hardware, they are vulner-
cess. Some cardiac catheterization laboratory physiologic
able to viruses and other malware. Systems should be
recorder/monitor systems (see below in this chapter) have
configured to prohibit installation of unauthorized applications
the capability to record device usage and generate reordering
on any hardware. They should also, ideally, not have either a
lists. Other options include computerized device storage cab-
web browser or an e-mail client installed. Ideally, systems
inets, which can be interfaced with computer systems to tabu-
should be isolated from the Internet. However, the challenge
late device usage. These systems are expensive and, while
in this area is that Internet-based connections are needed for
useful, are imperfect because they can easily be defeated by
remote monitoring and servicing of the system. Thus, the
user errors. Thus, their value to the administrative operation of
challenge is to provide remote monitoring and servicing access
a laboratory facility is tightly linked to the rigor of the operating
while excluding other potentially harmful Internet traffic.
procedures employed.
Ancillary Equipment Storage EQUIPMENT CONSIDERATIONS

Cardiac catheterization suites acquire numerous small portable System Integration
equipment items including ultrasound imaging machines, pres- A cardiac catheterization laboratory has two major equipment
sure and flow wire consoles, circulatory assistance machines, systems that work together.
and others. These machines should be stored in a designated 1. The X-ray cinefluorographic unit
area in the suite outside of general circulation (not in the 2. The physiologic monitoring system
procedure rooms or in the corridors) where they can be appro-
priately supported (connected to line electrical power if neces- The majority of current physiologic monitoring systems
sary to maintain battery charge) and readily located when provide both physiologic signal conditioning and display func-
needed. The size of this facility is determined by the amount tions for monitoring and recording purposes and also provide
and type of equipment stored. report generation and data archiving. The X-ray system and the
physiologic monitoring system need to communicate with each
other to link angiographic and physiologic data into a single
HOSPITAL NETWORK CONSIDERATIONS procedure file and report document. The task of achieving this
General Considerations communication capability is variable. The X-ray system ven-
Cardiac catheterization laboratories transmit and receive large dors also furnish physiologic monitoring systems and archiv-
quantities of computer network traffic. X-ray images are trans- ing systems. In this case, the communication protocols are built
mitted to and recalled from the laboratory’s archive server. into the systems and should operate seamlessly. However,
[2/7/010/7:17:52] [13–18]
there are also vendors that furnish stand-alone physiologic reducing image noise, increases radiation exposure to both the
monitoring systems with varying angiographic image archiv- patient and to the laboratory clinical personnel. Consequently,
ing capabilities. These vendors, who compete with the X-ray it is important that the hospital’s radiological physicist and
systems vendors by offering alternative functionality capabil- radiation safety officer oversee the calibration of the X-ray
ities, must establish communications with the X-ray systems to system to assure that optimal image quality is being generated
transfer patient and procedure identifier data as well as angio- at the lowest input doses achievable.
graphic image data if that is included in the system capability.
Consequently, when selecting equipment, compatibility and
Unit Configuration Issues
communications between systems must be assured to avoid
A number of issues need to be considered when specifying the
operational problems.
configuration of an X-ray unit. These include detector size,
As discussed above, when selecting equipment, it is
image processing capabilities, procedure table capabilities, and
important to confirm that, in addition to communication
biplane configurations.
between the X-ray system and physiologic monitoring system,
working communications links can be established with the hos-
pital information system to enable transmission of registration Detector Size
and accounting data and publishing of completed procedure X-ray image detectors suitable for cardiovascular imaging come
reports. These capabilities should be specified contractually at in two sizes 20 20 cm and 40 30 cm. The 20-cm detectors are
the time the systems are ordered. square and have 1024 1024 pixel matrices. The 40 cm
detectors are rectangular and have 2048 1536 pixel matrices.
It is important to point out that flat panel detectors provide
X-Ray Cinefluorographic Unit multiple image magnification modes. However, as these are
General Imaging Considerations digital devices, generation of a magnified image (e.g., a 20-cm)
The X-ray cinefluorographic unit is the core equipment around detector generally offers three image sizes. This is achieved by
which the entire laboratory facility is based. The technology of using only the detector’s central pixels and stretching their
these units has matured, and all of the vendors currently in the display to a larger size, thus magnifying the image. This
marketplace offer systems that are capable of generating excel- stretching of pixels is accompanied by a commensurate
lent fluorographic images. X-ray generator and X-ray tube increase in X-ray input dose, maintaining a constant dose-
design have now become relatively uniform across vendors. area product, to reduce image noise that would otherwise
Imaging chains have now migrated virtually completely from become evident in response to pixel magnification. Forty-cen-
X-ray image intensifier/video camera systems to integrated timeter detectors generate rectangular images in the 40-cm
flat-panel detector systems. mode but change to square images in magnified modes.
A major advantage of the migration to flat panel detector Detector size is an important consideration that is based
imaging chains is detector uniformity making the earlier quest on the unit’s anticipated usage pattern. The 40-cm detectors
for “the best image intensifier” a thing of the past. X-ray system offer the ability to achieve a larger image field of view but do so
manufacturers frequently make claims that flat panel detectors at the cost of greater bulk that impairs ability to achieve
require smaller X-ray input doses. However, in practice, this extreme degrees of cranial and caudal skew. The 40-cm detector
turns out not to be the case. is ideal for imaging large areas of the peripheral vasculature
With current systems, X-ray image quality is much more and also imaging substantially enlarged hearts (e.g., imaging
determined by the interaction of X-ray input dose and image the left atrium and left ventricle in a patient with severe
processing software than by the actual hardware components. enlargement of both chambers due to chronic mitral regurgita-
Consequently, any quality current system should be capable of tion, or imaging the left ventricle and thoracic aorta in a patient
generating high-quality images if its X ray–generating system, with severe left ventricular enlargement due to severe aortic
dose-modulating system, and image processing algorithms are regurgitation secondary to a thoracic aortic aneurysm). Thus, a
optimally calibrated. Thus, if a system is generating poor 20-cm detector is ideal for a laboratory that will be doing
images, the fault is likely with calibration rather than with mostly coronary imaging in patients with normal sized or
defective components in the imaging chain. only moderately enlarged hearts. A 40-cm detector will prove
The impact of current image processing algorithms on to be more cumbersome for coronary imaging and may actually
image characteristics cannot be overstated. In fact, many qual- preclude achieving certain highly skewed projections but will
itative differences in default image appearance characteristics be ideal for imaging enlarged hearts with valvular disease and
between different X-ray manufacturers are actually attributable angiography of the peripheral vasculature.
to philosophical choices about the characteristics of an optimal
image. While the raw initial image data generated by different
X-ray vendors are quite similar, the final image displayed on Digital Subtraction and Table Stepping
the monitor is strongly influenced by image processing choices Peripheral angiography is facilitated by two additional capa-
such as contrast ratio, white compression, and edge enhance- bilities in addition to detector size—digital subtraction and
ment algorithms. Thus, ideally, the end user should collaborate table stepping. Digital subtraction is very valuable when imag-
with the X-ray system manufacturer’s imaging specialists to ing below the diaphragm or in the neck. It frequently permits
achieve the image quality that is optimal in the opinion of the acquisition of diagnostic quality images with smaller contrast
end user physician. (but not X-ray) doses. Thus, it is a valuable adjunct for periph-
It is important that the end user physician keep in mind eral vascular work. Table stepping is useful principally to
the tradeoff between X-ray input dose and image quality. It is follow a contrast bolus injection below the inguinal ligament
easy for an X-ray vendor, in response to a request for better to the feet. Thus, it is of value for assessing infrainguinal
image quality, to increase the X-ray input dose. This, while arterial anatomy.
[2/7/010/7:17:52] [13–18]
SETTING UP A CATHETERIZATION LABORATORY AND EQUIPMENT CONSIDERATIONS 17
3-D Rotational Imaging Physiologic Monitor, Recorder, and Database

Another recently introduced capability of current digital X-ray In the past decade, the physiologic monitor and recorder sys-
units is rotational 3-D angiography. This technique rotates the tem has undergone a major evolution. This component, which
gantry rapidly through a 1808 arc during a coronary injection descended from a multichannel oscillograph with analog signal
acquiring images of the opacified coronary artery in multiple conditioning preamplifiers and an optical strip chart recorder,
projections. CT-type reconstruction algorithms are applied to has evolved into a comprehensive cardiac catheterization lab-
the image data set to enable a 3-D reconstruction of the coro- oratory information system. These systems now function as a
nary anatomy. The potential of this technique is to enable digital recorder and monitor that also incorporate procedure
generation of a comprehensive anatomic assessment of a cor- logging, report generating, and database capabilities. The addi-
onary artery with a single contrast agent injection. tion of these capabilities is the logical development of the
progressive application of computer technology to what was
originally an analog device.
Biplane Configurations
A typical current system incorporates the basic pressure,
Biplane configurations have attributes that are of value in three
ECG and other signal acquisition, display, and recording capa-
circumstances.
bility. The system has logic to measure digital values that
1. Patients with renal insufficiency in whom it is desirable to characterize intracardiac and intravascular pressure wave
minimize contrast agent dose. In such patients, if an oper- forms as well as calculate valve pressure gradients, cardiac
ator is skilled and experienced at performing biplane cor- output, vascular resistance, valve orifice areas, and intracardiac
onary angiography, a biplane X-ray unit offers the shunt flows. In addition, the unit, through its user interface,
potential for substantial contrast agent dose reduction. It records a time-stamped procedure log of all procedure events
is possible that, in future, 3-D rotational angiography and tabulates devices used for reporting and for inventory
may supercede this particular indication for biplane maintenance purposes. The recorder system archives the phys-
angiography. iologic data from the procedure including all of the physiologic
2. Patients with complex congenital heart disease who signal data obtained during the procedure. In addition, the
require multiple contrast injections with images acquired system has the ability to generate a clinical procedure report
in multiple projections can also benefit from studies con- using its recorded data combined with physician interpretation
ducted on a biplane unit. of the angiographic image data.
3. On occasion, biplane fluoroscopy is an adjunct when The physiologic recorder/database interfaces with the
performing interventional cardiovascular procedures as it procedure room X-ray system to establish links between the
enables rapid switching between fluoroscopic views, procedure’s physiologic data and its angiographic images.
which is sometimes helpful when conducting a complex Some systems are comprehensive physiologic data and image
interventional procedure. archiving systems, while others do not archive the angio-
graphic data. If the latter type of system is employed, a separate
A biplane X-ray unit costs nearly twice the price of a
angiographic archive is needed.
single-plane unit. As mentioned above, it also requires provi-
Particular attention should be paid to the ability of the
sion of a larger procedure room. The investment is wasted if the
physiologic recorder/database to interface with and communi-
lateral imaging plane hangs unused in the corner of the proce-
cate with the X-ray system. The X-ray vendors supply physio-
dure room. Thus, the choice to specify a biplane unit should be
logic recorder/database systems that are specifically designed
made carefully considering the uses intended for the particular
to interface with their X-ray systems. In addition, there are
laboratory. Many multiple procedure room facilities will equip
third-party vendors that offer physiologic recorder/database
one room with a biplane unit scheduling patients with the
systems and compete with the X-ray vendors on feature
above-cited circumstances in it while equipping other rooms
complement. A given cardiac catheterization suite may contain
with single-plane units. Similarly, a multiroom facility may
X-ray equipment from more than one vendor but should have a
have a mixture of detector sizes in its different rooms using the
single physiologic recorder/database system if it intends to use
rooms with large detectors for patients with severe cardiac
the system for reporting. Thus, there is the potential depending
enlargement and for peripheral vascular procedures while
on the supplier of the physiologic recorder/database system to
doing straightforward coronary work in rooms with 25-cm
need to interface one X-ray manufacturer’s physiologic record-
detectors.
ing system with another X-ray manufacturer’s X-ray unit.
Given the potential for conflicts, considerable planning must
Display Monitor Configuration be conducted to configure a blended manufacturer system. This
The configuration of the monitor displays is an important should include contractual guarantees to achieve full interoper-
ergonomic feature of procedure room design. The ceiling- ability of all linked systems.
suspended monitor system should facilitate positioning the
monitors where they can be easily viewed from all possible
catheter entry site locations. Ceiling-suspended monitor sup- Database Servers
port systems are available, which will support up to eight The heart of the laboratory information system is its database
monitors. As a general rule, one should design for two mon- server(s) (2). Depending on vendors used and configurations,
itors (live and roadmap) for each X-ray plane. In addition, the the laboratory may have a single archive server that stores both
monitor configuration will include the monitor for the physio- angiographic image data and physiologic monitoring and
logic recorder/monitor system for monitoring the patient dur- report generation data. Alternatively, these functions may be
ing the procedure. Consideration should be given to whether supported by physically different servers depending on the
additional monitors should be provided for display of other particulars of equipment configuration and vendors supplying
data such as intracardiac and intravascular ultrasound images. the systems. It is important to point out that these servers do
[2/7/010/7:17:52] [13–18]
not need to be located physically in the laboratory suite pro- access in which anatomy can be very variable with respect to
vided that they have connections with sufficient bandwidth as external landmarks. This unit is also of value in assessing
discussed above. In fact, it is likely preferable that these servers vascular access sites following catheter removal in case there
be located in the hospital’s larger computer facility where they is concern about vascular integrity.
can be maintained by dedicated computer support staff rather Intravascular and intracardiac ultrasound imaging is a
than be an item that requires periodic attention from the valuable adjunct to a variety in invasive and interventional
laboratory’s clinical staff (who are less likely to be adept at procedures. Since this imaging is used to support intracardiac
these functions). The same considerations regarding Internet catheter and device manipulation, these machines are ideally
connections described for the X-ray systems and monitoring interfaced with a monitor mounted on the main monitor boom
systems apply to the database servers. so that the operators have ready access to both ultrasound and
Database server capacity is an important consideration fluoroscopic images in the same location.
with financial implications. The server system is generally a
combination of three units. The server itself with an associated
redundant array of inexpensive disks (RAID) storage system
Guide Wire Pressure and Flow Velocity
provides online storage available immediately. A server “juke
Transducers
These devices are used for assessment of intracoronary pres-
box,” which contains an array of mountable removable media
sure and flow for measuring fractional flow reserve and coro-
data storage units, provides near-line storage generally avail-
nary vasodilator reserve. The devices are driven by portable
able automatically within several minutes of a request. Offline
consoles that need to be interfaced with the physiologic record-
storage is removable media not stored in a “juke box” that must
ing system. This constitutes one of the requirements for proce-
be physically mounted by an operator in a server drive in
dure table pedestal input and output connections other than
response to a request. Data in offline storage require variable
the standard pressure transducer input connections. These
amounts of time for retrieval. When specifying a database
connection capabilities must be designed into the table pedestal
server configuration, there is an obvious financial tradeoff
and cabled appropriately at the time of construction and
between the amount of online and near line capacity and the
installation.
cost of the system. These are functionality considerations that
should be made at that time of system specification.
REFERENCES
Ancillary Diagnostic Equipment: Provision, 1. Bashore TM. (chair) Task Force, American College of Cardiology/
Integration Society for Cardiac Angiography and Interventions Clinical Expert
Consensus Document on Cardiac Catheterization Laboratory Stan-
Ultrasound
dards. J Am Coll Cardiol 2001; 37:2170–2214.
A cardiac catheterization laboratory may employ a variety of 2. American College of Cardiology Foundation. Available at: http://
ultrasound equipment. This will entail portable consoles that www.cathkit.com/Cathkit/default.aspx.
drive transducers used for a variety of purposes.
A small duplex ultrasound machine is very useful for
assisting with vascular access, particularly internal jugular
[2/7/010/7:19:6] [19–27]
Radiation safety principles in the cardiac

catheterization laboratory
Thomas M. Bashore
INTRODUCTION digital converter, and the digital message is displayed on the

According to the most recent National Council on Radiation computer monitor.
Protection and Measurements report (1), Americans were In the flat panel system, the light photons emitted from
exposed to seven times as much ionizing radiation in 2006 the cesium iodide crystal strike a series of thin-film transistors,
than in the early 1980s. While much of this increase is the result rather than charge-coupled device silicon squares. The electron
of computed tomography (CT) and nuclear imaging, not an current produced is proportional to the number of light pho-
insignificant amount occurs in the cardiac catheterization lab- tons that hit each thin-film transistor. The video signal that
oratory. The general public has become increasing aware of the emerges reflects the amount of energy at each thin-film tran-
potential hazards of ionizing radiation (2). As a consequence, sistor and its location much like the charge-coupled device
the use of as-low-as-reasonably-achievable (ALARA) has chip. The signal is digitized (converted from analog to digital)
become a mantra for all healthcare providers to more carefully on the flat panel itself, and the digitalized video signal is then
examine the use of medical radiation. A recent review from the sent directly to the computer monitor for display.
American Heart Association Science Advisory committee has It is important to understand a couple of these steps in
further emphasized the importance of paying attention to more depth to appreciate where radiation exposure might
radiation effects in cardiac imaging (3). In general, the effects occur and how to best provide protection from it. The X-ray
of ionizing radiation have been a much greater focus for the tube (Fig. 3.2) is a glass enclosed vacuum tube with a cathode
radiologist than the cardiologist. This overview will very filament or coil inside a focusing cup. This filament emits
briefly discuss the manner in which X-ray images are created electrons when heated (thermionic emission). The temperature
to better understand the means by which a reduction in X-ray is >30008F. The focusing cup is negatively charged to help
dose may be accomplished. The radiobiology of ionizing radi- direct the emitted electrons toward the spinning anode where
ation will then be addressed to provide some insight into the either a large or a small focal spot area is targeted. The small
consequences of X rays on biologic tissue and how the effects focal spot provides better spatial resolution while the larger
can be minimized. focal spot is used to provide more X rays for imaging larger
patients. The number of electrons that leave the cathode and
head toward the anode is referred to as the mA (milliampere) of
THE BASICS OF X-RAY IMAGING IN THE CARDIAC the system. These electrons are encouraged to jump to the
CATHETERIZATION LABORATORY anode by applying a voltage potential across the X-ray tube.
Figure 3.1 outlines the basic operation of the digital X-ray This voltage potential is the kVp (peak kilowatts of voltage).
systems using a conventional image intensifier compared The anode rotates to distribute the heat that would otherwise
with a flat panel system. The two systems have several com- be striking a single spot. About 99% of the kinetic energy of the
monalities. In both systems, the flow of energy begins in the projectile electrons is converted to heat, so this is a very inef-
generator where electrons are sent to the X-ray tube and ficient system. Adding filtration increases the average energy of
converted into X rays. These X rays are emitted from the the X-ray beam by getting rid of the lower-energy X rays—a
X-ray tube and diverge quickly as they are beamed through process called beam hardening. Filtration thus improves the
the patient toward either the image intensifier or the flat panel image and reduces the radiation emitted. The beam can also be
receptor. A grid filters out stray X rays and only allows direct X shaped by collimators.
rays to hit a cesium iodide crystal, where they are converted to X rays are produced when the tungsten in the anode is
light photons. struck by the projectile electrons in two different ways (Fig. 3.3).
In the traditional image intensifier, these light photons Bremsstralung or braking X rays are emitted as a spectrum of
strike a photocathode and are converted back to electrons and energies when the projectile electron approaches the tungsten
accelerated toward an output phosphor by a voltage potential atom, slows down, and changes its path. Most diagnostic X rays
across the image intensifier of about 25,000 V. These electrons are formed by this method. If the projectile electron hits an inner
strike the output phosphor and produce a light image. The shell electron of the tungsten atom and an outer shell electron
X-ray image can be seen by looking directly at the output then hops inward to fill the vacant spot, the excessive energy is
phosphor. From there the image is captured on a silicon released as a characteristic X ray. These have discrete energies
light-sensitive charge-coupled device chip and converted to a and do not produce a continuous spectrum like the Bremsstra-
video signal. The video signal includes the amount of light that lung variety. As shown in Figure 3.3, the kVp across the X-ray
struck each tiny silicon square on the chip and the precise tube determines the maximal X-ray energy emitted. If the kVp of
location of that square. This signal is then sent to an analog to the system is increased, the amplitude of the emission spectrum
[2/7/010/7:19:6] [19–27]
Figure 3.3 The X-ray spectrum and the iodine absorption curve.
The emitted X rays have a wide range of energies, the maximal of
which is directly related to the kVp. Most X rays are produced by the
Bremsstralung method and only a few are characteristic X rays. An
iodine absorption spectrum is shown overlaying these energies. As
Figure 3.1 Basic components of the digital X-ray systems. The the X-ray energy increases, the iodine absorption drops until the K
classic system with its image intensifier (II) and video pick up system edge of the iodine is reached and suddenly there is a great deal of
is shown on the left with the flat panel system on the right. In both absorption of the emitted X rays (the K edge).
systems the X-ray generation is similar. Once X rays penetrate the
patient, the image intensifier and the flat panel detector have grids to
filter the X rays that pass to the input surface cesium iodide crystal
where light photons are generated. In the classic system, these light
photons release electrons in the image intensifier that are accel- shifts upward and to the right. If the kVp is kept the same but the
erated to an output phosphor where a light image appears. This mA of the system is increased, there is no right shift, but a major
image is captured using charge-coupled devices in the video cam- shift in the emission amplitude.
era, a video signal is generated and the data subsequently digitized If the absorption spectrum of iodine is superimposed
and displayed. In the flat panel system, the light photons trigger an over the X-ray emission spectrum (Fig. 3.3), it becomes clear
electrical signal from a thin-film transistors array and the video why iodine is used as an X-ray contrast agent—it absorbs the
signal is generated, digitized on the panel and the digital signal bulk of the energies emitted. As the intensity of the emitted
sent to the monitor for display. X rays increases, the iodine initially absorbs less and less of the
X-ray energies, then suddenly it absorbs most of them. This
spike in iodine absorption of X rays is due to the fact that the K
shell electron of iodine has an affinity for absorption of these
particular X-ray energies; this is sometimes referred to as the K
edge of iodine.
What is the relevance of all this? Understanding these
issues helps the cardiologist understand how changes in the
physical energy of the X-ray system affect the eventual image.
To obtain the optimal image, the X-ray system is constantly
trying to achieve a balance among all these factors by use of an
exposure equation: kVp mA pulse width. For instance, when
the kVp is increased, the iodine will absorb less of the high-
energy X rays emitted. This results in overpenetration of the
iodine column and less contrast between the desired image and
the background. Too high a kVp, thus washes out the image.
Keeping the kVp optimal and increasing the mA of the system
produces better images, but at the cost of more energy and
more radiation exposure. Increasing the pulse width results in a
blurred image of the beating heart. Thus, it is difficult to
increase either the kVp or the pulse width too much in acquir-
Figure 3.2 Operation of the X-ray tube and generator. A voltage
ing images in the catheterization laboratory as the images
potential is set up across the X-ray tube by the generator (maximal
voltage ¼ kVp) and electrons are sent to the cathode in the X-ray would either be washed out or blurred. So when a greater
vacuum tube. These cathode electrons heat the coil, “boil off,” and dose of X ray is needed, the principle way to increase the dose
jump to the anode because of the voltage potential. The number that is to increase the number of X rays sent (increase the mA).
leaps off is the mA of the system and is directly related to how many The X-ray quantity is the number of X rays produced.
X rays are generated. The higher the kVp, the greater the strength of When the mA is doubled, the number of X rays produced is
the X rays emitted. Collimators shape the beam and filters absorb doubled—a 1:1 relationship. The change in quantity is propor-
low-frequency X rays. The system constantly tries to fulfill the tional to the square of the ratio of kVp, however. In other
exposure equation (mA kVp pulse width) to optimize the image. words, if the kVp is doubled, the quantity of X rays increases by
a factor of 4—another reason to keep the kVp as low as possible
[2/7/010/7:19:6] [19–27]
RADIATION SAFETY PRINCIPLES IN THE CARDIAC CATHETERIZATION LABORATORY 21
for good image contrast. The X-ray intensity striking an object, the energy, the shorter the wavelength. As a generality, low-
such as a patient, also varies inversely by the square of the energy electrons interact with atoms, moderate energy elec-
distance from the X-ray target—the inverse square law. Thus trons interact with electrons and high-energy electrons interact
doubling the distance between the X-ray source and the patient with nuclei.
reduces the intensity of the X rays by a factor of 4. X-ray energy below about 10 keV interact with matter by
X-ray quality is similar to the penetrability or the ability of coherent or classical scattering (Fig. 3.4A). In this situation the
the X rays to penetrate through tissue. Distance and mA do not X ray interacts with an atom and it becomes excited, immedi-
affect this, but, as outlined above, the kVp does. High-energy X ately releasing the same amount of excess energy but in a
rays penetrate tissue more than low energy. Filtration of the different direction from the incident X ray. This is simply
beam to get rid of low-energy X rays improves the quality. scatter of energy with no ionization.
Barriers, such as a lead apron, reduce the penetration of X rays. Both high- and low-energy X rays can interact with the
The ability of these barriers to do their job is measured in the outer shell electrons of an atom and not only ionize the atom,
half-value layer (HVL). The HVL is the thickness of absorbing but can reduce its energy—the Compton effect. This occurs
material necessary to reduce the X-ray intensity to half its when the X ray knocks the outer shell electron from the atom
original value. (Fig. 3.4B). This scatters the X ray in a different direction and
the ejected electron is referred to as a Compton electron. Both
the scattered X ray and Compton electron may go on to interact
with other atoms before they lose their energy. Scattered X rays
X-RAY INTERACTION WITH MATTER provide no useful information and can produce a uniform haze
While X rays interact with matter in five basic ways: coherent on the image, resulting in loss of image contrast. They also are a
scattering, the Compton effect, the photoelectric effect, pair major source of radiation hazard for both the patient and the
production and photodisintegration, only the first three inter- operators.
actions are important in the cardiac catheterization laboratory. Another interaction is the photoelectric effect wherein the
Electromagnetic radiation tends to interact with structures that X-ray photon is absorbed by an inner shell electron, ejecting the
are similar in size to the wavelengths of the radiation. X rays electron—now called a photoelectron (Fig. 3.4C). These photo-
have very short wavelengths (from 108–109 m). The higher electrons may now interact with other atoms increasing scatter.
Figure 3.4 X-ray interaction with matter. Shown is the classic Bohr atom. (A) Coherent or classical scatter. In this situation the energy of the
incident X ray is diverted but loses no energy. The wavelength of the scattered X ray is the same as the incident X ray. (B) Compton scatter. In
this situation the incident X ray knocks out the outer shell electron and ionizes the atom. The wavelength of the scattered X ray is greater than
the incident X ray and a Compton electron is ejected. Both can now interact further with other atoms. (C) Photoelectric scatter. In this scenario
the incident X ray is totally absorbed while knocking out an inner shell electron. The incident X ray disappears and the ejected electron is
called a photoelectron and can now interact with other atoms.
[2/7/010/7:19:6] [19–27]
Other interactions include pair production wherein an X ray Table 3.1 Cell Type and Radiosensitivity
interacts with the nuclear force field and two electrons of
Radiosensitivity Cell type
opposite charge are created. This interaction is not important
in diagnostic X-ray imaging. Nor is photodisintegration, where High Lymphoid tissue
high-energy X rays are directly absorbed by the nucleus and the Bone marrow
excited nucleus releases a nucleon or other fragments. Gonads
Intermediate Skin
The X-ray images you see is almost entirely created by
Gastrointestinal tract
how many X rays go through the patient unscathed and how Cornea
many interact with matter by either the Compton or the pho- Growing bone
toelectric effect and are thus absorbed. The interaction of the Kidney
X rays and tissue is proportional to the mass density of the Liver
tissue. Therefore bone absorbs more X rays than the lungs, and Thyroid
one must increase the X-ray dose to penetrate bone compared Low Muscle
with the penetration of lungs. Brain
In the end, only about 1% of the incident X rays from the Spinal cord
X-ray tube make it through all this and reach the image Source: From Ref. 4.
intensifier or flat panel. Increasing the kVp of the system
reduces the number of direct Compton scatter interactions
but increases the photoelectric effect and the Compton scatter- most radiosensitive. Tissue is also more sensitive when irradi-
ing due to that. The result is that there is more X-ray scatter at ated in the oxygenated or aerobic state. Hyperbaric conditions
high kVp levels as mentioned above. have been used in radiation oncology to take advantage of this.
Grids are placed on the input of the image intensifier or DNA injury from radiation can be either direct or indirect
flat panel to reduce imaging of scattered X rays. They improve (Fig. 3.5A). Direct damage to the DNA backbone occurs when
image contrast, but the more the scatter, the higher the dose of there is ionization of a backbone molecule and a break occurs.
X ray needed to obtain a satisfactory image.
At the output of the image intensifier or flat panel, an
automatic exposure control system analyzes the image produ-
ces and provides immediate feedback to the generator on
whether the exposure equation has been fulfilled. The genera-
tor pulses the X rays and the fewer the frames/second used the
lower the overall dose. Magnification occurs when only a
central portion of the input face of the image intensifier is
used or when the source-to-image distance (SID) is increased
and many X-ray photons are lost because of the divergence of
the X rays from the X-ray tube. Magnification always results in
the need for a greater dose, therefore, and the avoidance of
magnified views helps reduce X-ray radiation exposure.
RADIOBIOLOGY
It is clear that X rays are harmful to tissue. This is due to the
interactions of X rays at the atomic level and the resulting
ionization of the atoms or the deposition of the energy into the
tissue. Deposited energy can result in a molecular change.
Ionization changes the chemical bonding properties of the
atom and can result in the molecule breaking up or the atom
relocating within the molecule. This can result in the molecule
no longer functioning and can impair or kill the cell.
The target radiosensitive molecule in the cell is the DNA.
It is composed of a backbone of alternating segments of a sugar
(deoxyribose) and a phosphate. Attached to the backbone is one
of four nitrogenous bases (adenine, guanine, thymine or cyto-
sine). The molecule is strung together in the familiar double
helix ladder with the backbone of alternating sugar-phosphate
molecules and the ladder rungs of the base pairs (adenine-
thymine or guanine-cytosine). Mitosis is defined when the cell Figure 3.5 DNA Injury. (A) Direct and indirect injury. DNA can be
divides, and this is when the cell is most susceptible to radia- directly affected by the X-ray energy or may be injured by free
tion damage. Organs that are the most susceptible to radiation radical formation. Two-thirds of injury is from free radicals.
injury are therefore those that divide the most frequently (4) (B) Types of DNA injury. The injury to the DNA backbone can be
(Table 3.1). These tend to be bone marrow cells, lymphoid a single break, a double break, a break and subsequent cross-linking
tissue and gonads. Stem cells are more radiosensitive than or may be injury to the amino acid rungs of the DNA. Single breaks
mature cells. Similarly, younger tissue, those with the highest and noncontinguous breaks usually heal quickly. Source: From Ref. 5.
metabolic rate and those with a high proliferation rate are the
[2/7/010/7:19:6] [19–27]
Indirect injury occurs when the X ray interacts with water ronmental agents. The exact role of hormesis in reducing the
producing electrically charged H2Oþ þ e. H2Oþ is an ion risk of radiation has not been established and the concept
radical since it contains an unpaired electron in its outer shell. remains controversial.
Ion radicals have a very short half-life of 1010 sec. They decay
to form a free radical by interacting with another water mole-
cule. RADIATION SAFETY ISSUES
þ þ
Definitions on how to describe the amount of radiation dose
H2O þ H2O ? H3O þ OH that is delivered to the tissue have varied over the years. The
The free hydroxyl radical possesses nine electrons, so one radiation absorbed dose (rad) is a measure of the energy absorbed
of them is unpaired. It is highly reactive and can diffuse a short per unit mass by an organ. It is expressed in units of mGy
distance to reach DNA. About two-thirds of DNA injury is by (milliGray). The absorbed dose depends on the absorbing
free radical injury and about one-third by the direct effect of material and the photon energy. One rad ¼ 10 mGy. The
X rays. effective dose (rem) strives to reflect the overall result of being
When a DNA strand break occurs (Fig. 3.5B), it tends to exposed to ionizing radiation. It is an attempt to represent the
be rapidly repaired using the opposite strand as a template. If amount of whole body radiation that occurs during radiation of
the repair is incorrect (misrepair), it may result in a mutation. If only a portion of the body, such as in the catheterization
both strands are broken but the places where the breaks occur laboratory. It is expressed in mSv (milliSieverts). The effective
are separated up and down the strand from each other, then dose is derived from simulations of radiation exposure using
healing tends to occur quickly. If both strands are broken in a mathematical models plus the radiation weighting factor (for
similar place and the DNA breaks into two pieces, then the cell X rays it is 1.0) plus tissue-specific weighting factors. One
dies, a mutation is created or carcinogenesis occurs. Other types rem ¼ 10 mSv.
of DNA injury include a DNA break and then cross-linking Table 3.2 outlines the estimated effective dose for a vari-
with another piece of DNA or injury to the amino acid rungs of ety of cardiovascular imaging procedures to order to put car-
the DNA structure (rung breakage). These latter are also diac catheterization radiation dosage in perspective. We live on
referred to as point mutations and can result in incorrect a radioactive planet and are exposed to cosmic radiation as well
code being transferred to daughter cells. DNA damage can as man-made sources. Background radiation includes about
result in abnormal metabolic activity and uncontrolled growth 82% from natural resources and 18% from man-made radiation.
of the tissue. For a chromosome break to be obvious a large Of the background radiation, the bulk (55%) comes from earth
amount of DNA must be destroyed. radon (a particle) exposure, with the average total background
A deterministic radiation injury is said to be present when radiation exposure in the United States around 3.6 mSv. If a
a certain number of cells of an organ die following radiation routine PA chest X ray results in 0.04 mSv of exposure, we
injury, such as a skin burn. These types of injuries are dose therefore receive the equivalent of about 90 chest PA X rays per
dependent, and there is a threshold when the effects become year from background radiation. Patient exposure in the car-
obvious. A stochastic radiation injury results in radiation can- diac catheterization laboratory amounts to about 7 mSv for a
cers or genetic effects and can occur with only a single DNA diagnostic catheterization (twice background) and around dou-
break. It, therefore, has no threshold dose, though the risk of a ble that for a coronary interventional procedure.
break increases linearly as the dose increases. Hormesis is a term To gauge how much radiation exposure the operator
used to describe the concept that a little radiation may be good receives, a radiosensitive badge of some sort is used—either a
for you (Fig. 3.6). The explanation is that a little radiation may thermoluminescent dosimeter (TLD) or by an optical simulated
stimulate hormonal or immune responses to other toxic envi- luminescent (OSL) badge. TLD badges utilize a LiF crystal to
absorb X rays and release light photons proportional to the
Table 3.2 Representative Values for the Effective Dose Estimates

for Various Cardiovascular Imaging Studies
Representative
effective
Cardiovascular study dose (mSv)
Background radiation 3.6
Chest X ray (PA and lateral) 0.04
Diagnostic cardiac catheterization 7
Percutaneous coronary intervention 15
Coronary calcium computed tomography 3
Coronary CTA (ungated) 15
Coronary CTA (triggered) 3
Myocardial perfusion
Sestamibi (99mTc 1-day protocol) 9
Sestamibi (99mTc 2-day protocol) 13
Thallium-201 41
Figure 3.6 The concept of hormesis. A radiation dose-response F-18 FDG 14
curve is shown suggesting a reduced response at low doses of Rubidium-82 5
radiation exposure. Source: From Ref. 6. MUGA (99mTc-labeled RBCs) 8
Source: From Refs. 3, 7, and 8.
[2/7/010/7:19:6] [19–27]
RISKS FROM IONIZING RADIATION

The health effects from radiation has been the subject of much
debate over the years. The Committee on the Biological Effects
of Ionizing Radiation (BEIR), an arm of the National Academy
of Sciences recently published the seventh in a series of reports
on the issue (BEIR VII) (15). In it they support the concept of a
“linear-no-threshold” model for cancer risk. This assumes that
even the smallest dose of ionizing radiation carries a tiny, but
potential risk. The committee pointed out the risk is a function
of age at exposure, the known gap between exposure and
manifestation of disease and whether the response was abso-
lute or relative. There is a latent period after exposure before
the risk of cancer is manifest. The magnitude of this risk is
strikingly low, however, and for most occupational radiation
Figure 3.7 The dose-area-product (DAP) and the IRP. The DAP is exposure the chances of dying in a year is about equivalent to
calculated from the output of the X-ray tube by use of an ionizing the risk of air travel.
chamber. It represents total X-ray energy emitted from the X-ray Linear energy transfer (LET) is a measure of the rate at
tube. The IRP is determined by assuming the skin of the patient is which energy is transferred from ionizing radiation to tissue.
150 mm from the isocenter of the patient’s body. Abbreviation: IRP, Alpha particles (such as radon) have a high value for LET (100
interventional reference point. keV/mm) while diagnostic X rays have a low value (3 keV/mm).
Low LET results in less damage to the tissues. The BEIR VII
report defines low doses in the range of near zero to about 100
mSv of low LET radiation (15). Using data from the Japanese
atomic bomb survivors, about 60% of survivors were exposed to
less than 100 mSv. On average, assuming a sex and age distri-
X rays absorbed when heated. The OSL badge contains an bution similar to the U.S. population, 1 in 100 persons would be
aluminum oxide doped with carbon and releases light in pro- expected to develop cancer from exposure of 100 mSv while 42
portion to the X rays absorbed when later struck with a laser. in 100 would develop cancer from other causes. At doses less
Differing filters mimic attenuation one might expect for shal- than 100 mSv, the BEIR VII committee concluded that the risk
low, lens or deep exposure. Proper use of radiation badges is to would be extremely low and would drop linearly with no
wear them exposed on the thyroid collar and under the lead threshold (15) as demonstrated in Figure 3.8. The BEIR commit-
apron at the waist. Unfortunately, few wear the waist badge, so tee did not feel there was evidence to support any benefit from
most data related to exposure is derived from the thyroid collar low doses of radiation—the hormesis described above.
badge.
Radiation to the patient is difficult to quantitate. It can be
estimated using the DAP (dose-area-product) and/or the inter-
ventional reference point (IRP). These concepts are shown in
Figure 3.7. The DAP is measured by means of an ionization
chamber placed at the output of the X-ray tube. It is the
absorbed radiation dose to air kerma multiplied by the X-ray
beam cross-sectional area at the point of measurement. It is
expressed in Gy/cm2. It is an attempt to estimate how much
X-ray dose exited the X-ray tube during the study. An average
value for a diagnostic catheterization is about 45 Gy/cm2 and
for an interventional procedure about 75 Gy/cm2 (9), though
recent values of much higher values have also been reported
(10). Values of DAP over 300 Gy/cm2 are more likely to result
in skin injury to the patient (11). In pediatrics, the values have
been much lower than in adults, with diagnostic DAP values of
7.77 Gy/cm2 and even PDA closure values of only 23.21 Gy/
cm2 (12). Because stochastic effects are cumulative, some have
suggested the DAP be a permanent part of each patient’s record
(13). It is only a surrogate, though, for the total amount of
radiation actually delivered to the patient.
The IRP is an attempt to figure out how much dose was
delivered to the patient’s skin. It is determined by assuming the Figure 3.8 Linear risk of cancer. While there are both linear and
patient’s skin is about 15 cm from his isocenter (9,14). If one nonlinear models for cancer risk from radiation, the latest BIER VII
knows the distance from the X-ray tube to the isocenter, then consensus report suggest a linear model is appropriate for low-dose
the distance to the skin facing the X-ray tube is presumed 15 cm exposure. The graph, however, is derived from high-dose single
toward the X-ray tube. The delivered X-ray dose at that point exposures experienced in Japan and Chernobyl after nuclear acci-
can then be estimated from the DAP and the distance to the dents, and the ability to extrapolate to the low levels used in medical
skin (1/d2). If the calculated value is >4 Gy, it is likely the imaging remains controversial. Source: From Ref. 6.
patient may develop a skin rash (9).
[2/7/010/7:19:6] [19–27]
Table 3.3 Threshold Skin Entry Doses That Result in Skin Injury Table 3.4 Days of Life Lost as a Consequence of Occupation,
Disease, or Other Conditions
Dose Effect Onset
2 Gy (2000 mSv) Erythema 1 hr Condition Expected days of life lost
4–6 Gy Late erythema 10 days to 10 wk Being male (vs. female) 2800
7 Gy Hair loss 3 wk Heart disease 2100
10 Gy Atrophy, fibrosis 14 wk to 1 yr Being unmarried 2000
6–18 Gy Necrosis 10 wk to 1 yr One pack of cigarettes a day 1600
Unknown Skin cancer >5 yr Coal mining 1100
Source: From Refs. 9 and 17. Cancer 980
30 lb overweight 900
All accidents 435
Radiation worker 12
PATIENT RISKS Airplane crashes 1
As shown in Figure 3.1, as X rays leave the X-ray tube and Source: From Ref. 20.
traverse the patient, they are absorbed and scattered. When
only local tissue is irradiated, it takes a higher dose to produce
an adverse response, as opposed to the risk from whole body
irradiation. The effect on any particular organ is individual cell Despite these reassuring data, a recent consensus state-
death and shrinkage of that organ or tissue. It also takes a ment from the major American societies of physicians who
threshold level to produce a noticeable change in the organ’s work in the interventional laboratory environment sends an
function—a deterministic response to radiation. Once the alarm that the risk is simply not well defined (23). The group
threshold has been reached, then the severity of the adverse notes the epidemic of orthopedic issues in these environments,
response increases with dose. and sites case reports of increased cancers, particularly of the
The patient’s skin first greets X rays in the catheterization brain (24) and bone marrow (25). They point out that there is no
laboratory and receives the brunt of the radiation injury during shielding of the central nervous system, and that radiation
cardiac catheterization. Table 3.3 outlines the threshold doses exposure has been associated with neural tumors. The also
for skin injury and the timing when these injuries become suggest that cataract formation may be a stochastic and not
evident after exposure. Certain diseases appear to predispose deterministic effect, and that the dose upper limits may cur-
the patient to skin injury from radiation, including collagen rently be too high to protect. Finally they lament that many
vascular disease, diabetes mellitus, hyperthyroidism, ataxia interventional cardiologists may be exposed to over 30 years of
telangiectasia and prior exposure to radiation (9,18,19). radiation with no real data to know the actual risks involved.
The risk of a stochastic skin effect (cancer) is difficult to Table 3.5 outlines the suggested recommended dose
estimate, but the relative risk of 4:1 has been reported from limits from the National Council on Radiation Protection and
exposure of 5 to 20 Gy, 14:1 from 40 to 60 Gy, and 27:1 from 60 Measurement. Note that the annual maximal dose (on your
to 100 Gy (20). Stochastic effects are cumulative; though it is badge) is 50 mSv to reduce a stochastic effect and 150 mSv to
very unlikely patients would receive doses of this magnitude prevent a deterministic effect. Since stochastic effects are cumu-
from low radiation studies such as cardiac catheterization. lative, the maximal dose over a lifetime is 10 mSv your age or
The risk of stochastic effects to other organs in the patient a maximum of 500 mSv. During the average interventional
is incompletely known, but there may be a small but detectable procedure, you, as the operator, receive anywhere from 0.07 to
increase in solid tumors at doses as low as 100 mSv (21). The 0.31 mSv on the badge (28) (the equivalent of 1.5 to over 7 chest
general risk of a fatal cancer has been estimated at 0.004% to X rays of dosage). Unfortunately, there are a variety (up to six)
0.12% for each 10 mSv exposure (9). Newborns are estimated to different formulae being used to convert the data from the
be 10 to 30 times more sensitive to radiation and females appear badge into the effective dose, and the method used can result in
more susceptible than males (22).
Pregnancy is a special situation. Fetal risk is greatest
during the first trimester. The risk of leukemia has been Table 3.5 Recommended Radiation Effective Dose Limits
estimated at 0.06% per 10 mSv exposure. The risk of later Background radiation 3.6 mSv
adult cancer is unknown. Neurologic tissue appears to be at PA chest X ray 0.04 mSv
greatest risk and mental retardation after the atomic bomb Average occupational exposure during PCI 0.07–0.20 mSv
survivors was observed. Pregnancy is not an absolute contra- Maximal annual exposure limits
indication to cardiac catheterization. Shielding from direct Stochastic effects
X-ray exposure is relatively effective when the heart is imaged, Annual 50 mSv
and it has been estimated that less than 2% of the delivered Cumulative 10 mSv age in yr
dose scatters to reach the uterus (9). Deterministic effects
Annual
Eye 150 mSv
OCCUPATIONAL RISKS Skin 500 mSv
Occupational risks from radiation in the cardiac catheterization Embryo or fetus 2 mSv total to abdomen (0.5 mSv/mo)
laboratory are poorly understood, but by all measures they are Public exposure
quite low. One way to look at this is shown in Table 3.4 where it Annual
is suggested that a radiation worker can expect to lose only 12 Stochastic 1.0 mSv/yr
days from his lifespan compared with heart disease that would Deterministic 50 mSv/yr
result in 2100 days lost, for instance. Source: From Refs. 26 and 27.
[2/7/010/7:19:6] [19–27]
Table 3.6 Ways to Reduce Radiation Exposure in the Cardiac

Catheterization Laboratory
Patient specific
Vary views to minimize dose to any specific skin area
Shielding of gonads
Use of the interventional reference point to estimate skin
deterministic risk
To both the patient and operator
Proper collimation of the X-ray beam
Use of filters at the output of the X-ray tube
Keep the image intensifier as close to the patient as possible
(minimize the source-to-image distance)
Use the lowest framing rate possible
Use high-dose fluoroscopy only when absolutely necessary
Use pulsed fluoroscopy
Figure 3.9 Relationship of the X-ray source and the operator. In
Keep magnified views to a minimum
the left panel, the X-ray tube during a cranial LAO acquisition is
Use the minimum number of views
close to the operator. The dose one might receive from this position
Pay attention to angulated views
may be up to six times the dose when the X-ray tube is on the
Record the dose-area-product to look for ways to minimize dose
opposite side of the table in an RAO position shown on the right.
Operator specific
Source: From Ref. 16.
Remember: time/distance/barriers
Stay as far from X-ray source and patient scatter as possible
Use all available shielding, barriers and eyewear
Keep track of exposure by use of radiation badges
Pay attention to relationship of patient, the X-ray source and you
considerable variability in the estimated dose (29), further
adding to the uncertainty of these guidelines. This has con- Source: From Refs. 9 and 32–34.
tributed to the wide variations in the radiation doses to
operators that have been reported (28).
Exposure is highest on the left side of the body because of
REFERENCES
1. Medical Radiation Exposure of the U.S. Population Greatly
the relationship of the operator to the X-ray tube and patient Increased Since the Early 1980s. NCRP Publications. 3-3-2009.
(28). Indeed the location of the operator to the X-ray tube is 10-19-2009.
critical. In an LAO cranial view, the X-ray tube is closest to the 2. Lauer MS. Elements of danger–the case of medical imaging. N
operator and the dose received may be six times that from the Engl J Med 2009; 361:841–843.
308 RAO view where the X-ray tube is on the opposite side of 3. Gerber TC, Carr JJ, Arai AE, et al. Ionizing radiation in cardiac
the table (16) (Fig. 3.9). Studies performed by the radial method imaging: a science advisory from the American Heart Association
may result in a substantial increase in operator radiation expo- Committee on Cardiac Imaging of the Council on Clinical Cardi-
sure (30). ology and Committee on Cardiovascular Imaging and Interven-
Operator exposure comes from both scattered radiation tion of the Council on Cardiovascular Radiology and Intervention.
Circulation 2009; 119:1056–1065.
from the patient and directly from the X-ray tube (including
4. Bushong SC. Human biology. In: Bushon SC, ed. Radiologic
leakage). As a general rule, the closer you are to the X-ray tube Science for Technologists: Physics, Biology and Protection. 9th
itself, the greater your radiation exposure. Similarly, the greater ed. St. Louis: Mosby Elsevier, 2008:500–511.
the dose used, the greater your exposure. Thus, when the SID is 5. Bushong SC. Molecular and cellular radiobiology. In: Bushon SC,
wide, the divergence of X rays means that many X rays will not ed. Radiologic Science for Technologists: Physics, Biology and
strike the receiving image device and many will be lost. A wide Protection. 9th ed. St. Louis: Mosby Elsevier, 2008:520–533.
SID therefore requires more X-ray dose and results in more 6. Bushong SC. Fundamental principles of radiobiology. In: Bushong
exposure to both the patient and the operator. The use of SC, ed. Radiologic Science for Technologics: Physics, Biology and
magnified views also results in the need for more dose to Protection. 9th ed. St. Louis: Mosby Elsevier, 2008:512–519.
satisfy the exposure equation and more radiation to both the 7. Fazel R, Krumholz HM, Wang Y, et al. Exposure to low-dose
ionizing radiation from medical imaging procedures. N Engl J
patient and the operator as well. Pulsed fluoroscopy at rates
Med 2009; 361:849–857.
below 25 pulses/sec reduces dosage, but not at higher rates 8. Mettler FA Jr., Huda W, Yoshizumi TT, et al. Effective doses in
(31). Table 3.6 outlines the various means by which one can radiology and diagnostic nuclear medicine: a catalog. Radiology
reduce radiation to the patient and to the operator during 2008; 248:254–263.
cardiac catheterization. As one editorial pleaded: “Shielding 9. Hirshfeld JW Jr., Balter S, Brinker JA, et al. ACCF/AHA/HRS/
really works when you use it” (35). It is important to remember SCAI clinical competence statement on physician knowledge to
that not only are operators exposed over a long period, many optimize patient safety and image quality in fluoroscopically
patients will have multiple procedures using ionizing radiation guided invasive cardiovascular procedures: a report of the Amer-
over a lifetime, and stochastic risks are cumulative. Also recall ican College of Cardiology Foundation/American Heart Associa-
that obese patients require greater doses as well, and with the tion/American College of Physicians Task Force on Clinical
Competence and Training. Circulation 2005; 111:511–532.
epidemic of overweight people in this country, the average X-
10. D’Helft C, McGee A, Rainford L, et al. Proposed preliminary
ray dose per patient has gone up. diagnostic reference levels for three common interventional car-
In the end, remember the ALARA principle: (i) assume diology procedures in Ireland. Radiat Prot Dosimetry 2008; 129:
there is no absolutely safe dose of ionizing radiation; (ii) recall 63–66.
the smaller the dose, the less risk of an adverse event; and (iii) 11. Neofotistou V, Vano E, Padovani R, et al. Preliminary reference
incremental radiation doses have a cumulative effect. levels in interventional cardiology. Eur Radiol 2003; 13:2259–2263.
[2/7/010/7:19:6] [19–27]
12. Al-Haj AN, Lobriguito AM, Rafeh W. Variation in radiation doses 23. Klein LW, Miller DL, Balter S, et al. Occupational health hazards
in pediatric cardiac catheterization procedures. Radiat Prot Dos- in the interventional laboratory: time for a safer environment. J
imetry 2008; 129:173–178. Vasc Interv Radiol 2009; 20:S278–S283.
13. Bogaert E, Bacher K, Lemmens K, et al. A large-scale multicentre 24. Finkelstein MM. Is brain cancer an occupational disease of car-
study of patient skin doses in interventional cardiology: dose-area diologists? Can J Cardiol 1998; 14:1385–1388.
product action levels and dose reference levels. Br J Radiol 2009; 25. Matanoski GM, Seltser R, Sartwell PE, et al. The current mortality
82:303–312. rates of radiologists and other physician specialists: specific causes
14. Chida K, Kagaya Y, Saito H, et al. Evaluation of patient radiation of death. Am J Epidemiol 1975; 101:199–210.
dose during cardiac interventional procedures: what is the most 26. Radiation protection. In: Hall EJ, ed. Radiobiology for the Radiol-
effective method? Acta Radiol 2009; 50:474–481. ogist. 5th ed. Philadelphia: Lippincott Williams and Wilkins,
15. Monson RR, Cleaver JE, Abrams HL, et al. and Committee to 2000:234–248.
Assess the Health Risks from Exposure to Low Levels of Ionizing 27. National Council on Radiation Protection and Measurements.
Radiation. Beir VII: health risks from exposure to low levels of Recommendations on the limits for exposure to ionizing radiation.
ionizing radiation, 2006. Washington, D.C., 20001, National Aca- NCRP Report No. 116, Bethesda, Maryland, 1993.
demies Press, November 1, 2009. 28. Kim KP, Miller DL, Balter S, et al. Occupational radiation doses to
16. Balter S. Stray radiation in the cardiac catheterization laboratory. operators performing cardiac catheterization procedures. Health
Radiat Prot Dosimetry 2001; 94:183–188. Phys 2008; 94:211–227.
17. Wagner LK, Eifel PJ, Geise RA. Potential biological effects follow- 29. Chida K, Morishima Y, Masuyama H, et al. Effect of radiation
ing high X-ray dose interventional procedures. J Vasc Interv monitoring method and formula differences on estimated physi-
Radiol 1994; 5:71–84. cian dose during percutaneous coronary intervention. Acta Radiol
18. Koenig TR, Mettler FA, Wagner LK. Skin injuries from fluoro- 2009; 50:170–173.
scopically guided procedures: part 2, review of 73 cases and 30. Lange HW, von BH. Randomized comparison of operator radiation
recommendations for minimizing dose delivered to patient. AJR exposure during coronary angiography and intervention by radial
Am J Roentgenol 2001; 177:13–20. or femoral approach. Catheter Cardiovasc Interv 2006; 67:12–16.
19. Wagner LK, McNeese MD, Marx MV, et al. Severe skin reactions 31. Chida K, Fuda K, Saito H, et al. Patient skin dose in cardiac
from interventional fluoroscopy: case report and review of the interventional procedures: conventional fluoroscopy versus pulsed
literature. Radiology 1999; 213:773–776. fluoroscopy. Catheter Cardiovasc Interv 2007; 69:115–121.
20. Bushong SC. Late effects of radiation. In: Bushong SC, ed. Radio- 32. Bashore TM. Fundamentals of X-ray imaging and radiation safety.
logic Science for Technologists: Physics, Biology and Protection. In: Kern MJ, ed. SCAI Interventional Cardiology Review Book.
9th ed. St. Louis: Mosby Elsevier, 2008:549–567. Philadelphia: Lippincott Williams and Wilkins, 2007:91–102.
21. Pierce DA, Preston DL. Radiation-related cancer risks at low 33. Maeder M, Brunner-La Rocca HP, Wolber T, et al. Impact of a lead
doses among atomic bomb survivors. Radiat Res 2000; 154: glass screen on scatter radiation to eyes and hands in interven-
178–186. tional cardiologists. Catheter Cardiovasc Interv 2006; 67:18–23.
22. Brenner DJ, Elliston CD, Hall EJ, et al. Estimates of the cancer risks 34. Kuon E, Dahm JB, Empen K, et al. Identification of less-irradiating
from pediatric CT radiation are not merely theoretical: comment tube angulations in invasive cardiology. J Am Coll Cardiol 2004;
on “point/counterpoint: in x-ray computed tomography, tech- 44:1420–1428.
nique factors should be selected appropriate to patient size. 35. Balter S. Radiation shielding really works, when you use it.
against the proposition.” Med Phys 2001; 28:2387–2388. Catheter Cardiovasc Interv 2006; 67:24.
[14/6/010/13:1:45] [28–37]
Contrast agents
Michael C. Reed and Hitinder S. Gurm
INTRODUCTION cause less ventricular irritability than its ionic predecessors, but
Contrast agents used in the cardiac catheterization laboratory generally have a higher viscocity. Nevertheless, there appears
are effective in the radiographic visualization of the cardiac to be no increased risk of thrombotic events compared with
chambers and/or the coronary and peripheral vasculature. ionic agents (4). Additonally, nonionic LOCM are associated
However, these agents are associated with potential adverse with less nephrotoxicity and fewer allergic reactions compared
effects. Therefore, they have been modified over the years to with HOCM (5–7).
allow for the most effective imaging at the lowest potential There is some evidence that a nonionic contrast agent
toxicity. This chapter will discuss the types of contrast agents, with an osmolality similar to blood may be safer than LOCM.
their toxicity, and the evidence-based methods designed to Iodixanol (Visipaque) is a nonionic dimer which consists of two
prevent adverse effects. tri-iodinated benzene rings. It has an osmolality the same as
blood (290 mOsm/kg). Iso-osmolar contrast media (IOCM)
have a higher viscosity than HOCM and LOCM, but have
TYPES OF CONTRAST AGENTS been shown in some studies to cause fewer allergic reactions
Various formulations of contrast agents containing iodine have and are not associated with increased adverse coronary events
been used over the years to visualize the vasculature. All of (4,8). Iodixanol was associated with less nephropathy than
these formulations have similar concentrations of iodine, but ioxaglate in one randomized trial, and a meta-analysis of mul-
differ significantly in their structure, osmolality, and viscosity tiple small trials comparing LOCM and iodixanol suggests
(Table 4.1 and Fig. 4.1). iodixanol is less nephrotoxic (9,10). More recent randomized
First-generation, high-osmolality contrast media (HOCM) trials, however, have not shown a reduced incidence of con-
are ionic monomers, which consist of a single, negatively trast-induced nephrotoxicity due to iodixanol compared with
charged tri-iodinated benzene ring attached to sodium or other LOCM (11,12). However, multiple trials have shown a
another cation. As ionic media, they dissociate in solution and reduction in patient symptoms, allergic reactions, and organ
can cause cardiotoxicity related to calcium binding and repola- toxicity with LOCM or IOCM compared with HOCM.
rization changes during coronary angiography (2). Included in
this category are diatrizoate (Hypaque, Angiovist, Renografin),
metrizoate (Isopaque), and iothalamate (Conray). First- CHEMOTOXIC REACTIONS
generation agents have a sodium concentration similar to Contrast reactions can be divided into two major categories:
blood, but are hyper-osmolar relative to blood. Osmolality is chemotoxic reactions and hypersensitivity reactions. Hypersen-
typically >1400 mOsm/kg compared with 290 mOsm/kg found sitivity reactions, or allergic reactions, are idiosyncratic and
in blood. Both the ionicity and hyperosmolality of these agents independent of rate or volume of contrast infusion. Chemotoxic
contribute to significant volume shifts and to direct toxic effects reactions are related to the chemical properties of the agents
on the left ventricle and other organs. and are dose and rate dependent. These include volume over-
Second-generation, low-osmolality contrast media load, vasovagal reactions, cardiotoxicity, and nephrotoxicity.
(LOCM) were designed, in part, to minimize side effects related
to hypertonicity. Among the first such agents was the ionic
dimer ioxaglate (Hexabrix). Ioxaglate is a monoacidic double Volume Overload
benzene ring with a total of six molecules of iodine at the 2, 4, Most contrast media used in contemporary cardiac catheter-
and 6 positions on each ring. This agent can be used to visualize ization are hyper-osmolar relative to blood. First-generation
the vasculature at an osmolality of 600 mOsm/kg, twice that of agents have an osmolality which is up to six times that of blood.
blood, but less than half that of HOCM. Early studies con- Even LOCM have two to three times the osmolality of blood.
firmed that ioxaglate was associated with fewer side effects Rapid infusion of large amounts of contrast can result in large
compared with HOCM (3). fluid shifts from the extravascular to the intravascular space,
LOCM were further developed in the nonionic form. causing elevated filling pressures or even cardiogenic pulmo-
Monomeric forms of nonionic contrast agents are tri-iodinated nary edema in the supine patient. Use of LOCM or IOCM, or
with many hydrophilic hydroxyl groups and have an osmolal- minimization of contrast volume, or deferral of angiography
ity between 500 and 850 mOsm/kg, two to three times that of altogether may be considered in patients with evidence of
blood. These include iopamidol (Isovue), iohexol (Omnipaque), volume overload on history, physical exam, or direct measure-
iopromide (Ultravist), ioxilan (Oxilan), and ioversol (Optiray). ment of left ventricular end-diastolic pressure or pulmonary
Water soluble and without charge, the nonionic LOCM could capillary wedge pressure.
[14/6/010/13:1:45] [28–37]
CONTRAST AGENTS 29
Table 4.1 Comparison of Various Contrast Media

Iodine Sodium Osm (mOsm/ Viscocity
Generation Class Type of molecule Examples (mgI/mL) (mEq/L) kg-H20) at 378C
First High osmolar Ionic monomer Diatrizoate (Hypaque) 325–370 160 1797–2076 2.8–8.4
Metrizoate (Isopaque)
Iothalamate (Conray)
Second Low osmolar Ionic dimer Ioxaglate (Hexabrix) 320 150 600 7.5
Nonionic Iopamidol (Isovue) 350–370 2–5 695–844 8.1–10.4
monomer Iohexol (Omnipaque)
Iopromide (Ultravist)
Ioxilan (Oxilan)
Ioversol (Optiray)
Third Iso-osmolar Nonionic dimer Iodixanol (Visipaque) 320 19 290 11.8
Figure 4.1 Chemical structures of various contrast media.
Vasovagal Reactions from ionic contrast compounds. Observed electrocardiographic

It is very common for patients to experience a sensation of changes include sinus bradycardia, heart block, QRS widening,
flushing, warmth, or nausea with the injection of contrast QT prolongation, ST segment changes, and giant T wave inver-
media. This is typically mild, transient, and self-limited. Rarely, sion. Ventricular tachycardia and ventricular fibrillation may
this sensation can persist, and severe forms have been associ- also rarely occur, especially when injecting into a dampened
ated with hypotension and bradycardia. The mechanism of (ventricularized) coronary catheter. Transient left ventricular
vasovagal reactions from contrast is not known. These reactions systolic dysfunction and elevated left ventricular end-diastolic
should be distinguished from hypersensitivity reactions prior pressure have also been observed after contrast exposure.
to further injection of contrast media. Vasovagal reactions do Strategies to minimize contrast toxicity include using a ramped
not preclude further injection of contrast media, and slowing coronary injection, using the least amount of contrast possible
the rate of injection may ameliorate further symptoms. Because to fill the coronary artery, and counseling the patient to cough
patients rarely can have prolonged bradycardia or severe to clear contrast from the coronary tree in the event of an
hypotension related to increased vagal tone, it is advisable to arrhythmia.
disconnect the catheter from the power injector and reconnect
to the manifold to assess hemodynamics after a large, rapidly-
infused bolus of contrast media, such as with ventriculography.
Nephrotoxicity
A common and important chemotoxic contrast reaction is con-
trast-induced nephropathy (CIN). In some patients, particularly
Cardiotoxicity those with preexisting chronic renal insufficiency, contrast
Cardiotoxicity is less common with the introduction of non- media causes a significant rise of serum creatinine in the first
ionic LOCM and IOCM. The mechanism of direct cardiotoxicity few days after exposure. It typically begins in the first 12 to
is not well understood, but may involve the hypertonicity of 24 hours, peaks at two to three days, and resolves over five to
these agents or calcium-chelation of anions which dissociate seven days following exposure. The renal failure is typically
[14/6/010/13:1:45] [28–37]
nonoliguric, and the need for dialysis is highly uncommon vasodilatation. In comparison, increased levels of endothelin
(<0.1%) (13). and adenosine may promote vasoconstriction (29). This forms
The exact change in creatinine that defines CIN has the rationale for the experimental prophylactic drug theophyl-
varied from study to study, but the conventional definition is line, an adenosine receptor antagonist.
an absolute rise of serum creatinine of >0.5 mg/dL or a 25% Reduction in blood flow to the renal tubules and to the
increase from baseline serum creatinine. renal medulla may also be related to the hyperviscosity of
contrast media. The renal medulla is supplied by the vasa recta,
a collection of long, narrow-caliber vessels. Blood flow through
Epidemiology these small-diameter vessels may be particularly susceptible to
Nephrotoxicity from contrast media is a common cause of acute changes in viscosity.
renal failure in hospitalized patients. The incidence of CIN In addition to medullary hypoxic ischemia from
varies depending on the presence of risk factors, the type of decreased renal blood flow, there may be direct cytotoxic
contrast media, and the volume of contrast media. The inci- effects of contrast media on renal cells (24). There is some
dence is 1.2% to 1.6% (14,15). In patients with mild to moderate evidence that oxygen free radical generation may be increased
renal insufficiency (serum creatinine 1.5–4 mg/dL, the inci- after contrast media exposure. On the basis of this line of
dence is 4% to 11% (6,14,16,17). This increases to 10% to 40% in reasoning, the antioxidant acetylcysteine has been investigated
patients with mild to moderate renal dysfunction and diabetes for prophylaxis of CIN.
mellitus, and approaches 50% in patients with severe renal
insufficiency (serum creatinine of 4–5 mg/dL) (6,14,18). In a Risk Factors
Mayo Clinic registry of 7586 patients undergoing percutaneous There are modifiable and nonmodifiable risk factors that pre-
coronary intervention (PCI), the incidence of renal insufficiency dispose patients to CIN. These risk factors are likely additive.
(defined as a rise in creatinine of 0.5 mg/dL) was 3.3% overall Identification and modification of risk factors is crucial to
and 25% in patients with a baseline serum creatinine >2.0 mg/ minimizing nephrotoxicity.
dL (19). It should be noted that this study did not separate Age is an independent predictor of CIN in observational
causes of renal failure, and likely included some cases of renal studies and by multivariable analysis (30). The incidence of
failure related to atheroemboli or hemodynamic instability. CIN in patients greater than 70 years is approximately 11% (15).
This likely reflects the fact that elderly patients tend to have a
Prognosis lower glomerular filtration rate (GFR) and more medical
Very few patients with CIN require hemodialysis (<0.1%), but comorbidities. In addition, the elderly are more likely to have
CIN does increase hospital stay and is associated with worse multivessel coronary artery disease and to require PCI, which
short-term and long-term prognosis (13). In the Mayo Clinic translates to more contrast media exposure.
registry, the mortality was 22% in those patients with acute The most important predictor of CIN is the presence of
renal failure following PCI, compared with 1.4% in those underlying renal insufficiency. Chronic renal insufficiency
without renal failure (19). In a similar retrospective analysis (creatinine > 1.5 mg/dL or GFR < 60 mL/min/1.73 m squared)
of 9067 post-PCI patients, the one-year survival of patients with is consistently shown to increase the risk of CIN (15,19,31). In
renal failure was 70.3% versus 93.6% without renal failure (20). addition, the incidence of CIN increases dramatically with the
Although a creatinine rise of 0.5 mg/dL may not seem signif- severity of chronic renal insufficiency. This is especially true
icant, a retrospective analysis of 16,248 patients showed that when underlying renal insufficiency is due to diabetes mellitus.
even small rises in creatinine were associated with increased Inadequate renal perfusion in the setting of congestive
mortality risk (21,22). heart failure or hemodynamic instability is also associated with
increased risk for CIN. This is particularly true in the setting of
Pathophysiology a large anterior myocardial infarction or the need for intra-
The exact pathophysiology of CIN is not clear, although many aortic balloon counterpulsation (15,32). Anemia has also been
potential mechanisms have been implicated. Because of the identified as an independent risk factor in multivariable anal-
typically transient and self-limited nature of CIN and because ysis, perhaps related to decreased oxygen delivery to tubular
of the frequent presence of concomitant etiologies of chronic cells (33).
kidney disease, renal biopsy is rarely performed. Animal It is intuitive that nephrotoxic drugs would further
models have shown changes consistent with acute tubular increase the risk of CIN. Nonsteroidal anti-inflammatory
necrosis (ATN) (23,24). However, recovery from CIN is faster drugs (NSAIDs), aminoglycosides, and other nephrotoxins
than the two to three weeks expected after other types of ATN. should ideally be avoided. The use of ACE inhibitors is con-
This suggests less severe or less permanent damage to tubular troversial, and studies examining whether these medications
cells, analogous to that observed with myocardial stunning increase risk have had mixed results (34,35).
after an ischemic event. In addition, the fractional excretion of Multiple myeloma has been associated with increased
sodium (FENa) observed in CIN is often <1%, which suggests risk of CIN. This is particularly true with HOCM. The incidence
prerenal hypoperfusion, as opposed to frank tubular necrosis. is likely < 1.5% with newer agents. There may be an interaction
One proposed mechanism of contrast nephrotoxicity is between contrast media and light chains, promoting deposits in
renal artery vasoconstriction. A transient increase in renal the tubules. Volume depletion, in particular, tends to cause
blood flow, followed by a prolonged decrease has been tubular precipitation of light chains after contrast media expo-
observed in CIN (25). The outer medulla seems particularly sure. The dose and the type of contrast medium used also
vulnerable to damage from decreases in renal blood flow. There influences the likelihood of CIN.
is evidence for decreased nitric oxide (NO) and increased Cumulative risk prediction models have been designed
adenosine and endothelin levels in subjects exposed to contrast to help predict the risk in a given patient (32). Freeman et al.
media (2628). NO is necessary for renal artery and arteriole developed a method of calculating a maximum predicted
[14/6/010/13:1:45] [28–37]
CONTRAST AGENTS 31
Table 4.2 Prevention of Contrast-Induced Nephropathy

Intervention Dose Mechanism Proven efficacy? Recommended?
0.9% saline 1 L IV prior to procedure Dilution Yes Yes
Minimize contrast volume Biplane imaging, limited Yes Yes
injections, staged
procedures
Use of low-osmolar or Unclear Yes Yes
iso-osmolar contrast
Acetylcysteine 600 mg PO BID day prior Antioxidant Controversial Yes
and day of procedure
Sodium bicarbonate 3 mL/hr IV for 1 hr prior Alkalinization Controversial Consider
and 1 mL/hr IV for 6 hr
after procedure
Vitamin C Antioxidant No No
Fenoldopam, dopamine, Improved renal artery No No
prostaglandin E1, or perfusion
theophylline
Trimetazedine Anticellular ischemia No No
Hemodialysis Removal of contrast No No
Abbreviation: IV, intravenous.
contrast media dose (MRCD = 5 mL body weight (kg)/serum prevention. Recognition of patients at risk is the first step.
creatinine (mg/dL). Exceeding this threshold was determined Identification of risk factors, and measurement of serum crea-
to be the strongest independent predictor of CIN requiring tinine with estimation of GFR or creatinine clearance prior to
hemodialysis after PCI. contrast exposure is essential. Use of an alternative diagnostic
modality such as ultrasound or magnetic resonance imaging
Clinical Features when possible is optimal, especially in those patients identified
CIN typically begins in the first 12 to 24 hours after contrast to be at high risk. Nephrotoxic medications should be held,
media exposure. Creatinine usually peaks between 48 and when possible. In addition to these general tactics, there are
72 hours, and renal function typically improves over the next specific preprocedural and procedural methods which are
five to seven days. Renal failure is typically nonoliguric, mild, effective in minimizing the risk of CIN (Table 4.2).
and transient. The need for hemodialysis is only 0.8% in Hydration There are no prospective randomized trials
patients who develop CIN after PCI (31). Not surprisingly, comparing hydration with no hydration, but there is universal
the need for hemodialysis translates into a bad outcome, with consensus that hydration is beneficial. The efficacy of hydration
an in-hospital mortality 36% (versus 1% of patients without in minimizing CIN may be related to dilution of contrast. In
hemodialysis) and a two-year mortality of 81% (31). addition, there is evidence that hydration minimizes the reduc-
It is important to consider other causes of renal insuffi- tion of nitrous oxide in the renal circulation. In an early study
ciency after contrast exposure. These include, but are not lim- evaluating various methods of prophylaxis against CIN in
ited to renal hypoperfusion or even ATN in the setting of patients with chronic renal insufficiency, 0.45% saline was
congestive heart failure, volume depletion, or sepsis; atheroem- associated with a significantly lower incidence of CIN (11%)
boli; interstitial nephritis; or obstructive uropathy. Renal failure than 0.45% saline þ mannitol (28%) or 0.45% saline þ furose-
related to atheroemboli is highest in patients with peripheral mide (40%) (37). Mueller et al. established that 0.9% saline
vascular and/or abdominal aortic aneurysmal disease who reduced CIN by more than half compared with 0.45% saline in
undergo cardiac catheterization. Unlike CIN, it may have a patients who received angioplasty (38). Finally, two trials
delayed onset, days or even weeks after catheterization. Athe- helped establish that intravenous (IV) saline was superior to
roemboli are associated with blue toes, livedo reticularis, fever, oral hydration with salt tablets (39,40). The optimal dose of
hypereosinophilia, hypocomplementemia, and ghost-like clefts hydration is not clear, and regimens may need to be modified
on renal biopsy from vacated cholesterol crystals. Unlike CIN, a in patients with impaired left ventricular function or evidence
high percentage of patients with atheroemboli-related renal of preexisting volume overload on physical exam.
insufficiency require hemodialysis, and recovery of renal func- Isotonic sodium bicarbonate has been proposed as an
tion is minimal (36). alternative form of hydration for prevention of CIN. Alkalin-
Urinalysis in CIN may reveal either low or elevated FENa, ization may further protect from free radical–induced tubular
presumably depending on degree of prerenal hypoperfusion injury or may decrease the viscosity of contrast agents passing
versus actual tubular injury. Urine protein may be elevated for through the vasa recta of the kidney. A reduction in CIN
the first 24 hours after contrast media exposure. Biopsy is rarely (defined as a 25% increase in serum creatinine) was found in
performed. By and large, the diagnosis of CIN is a clinical a small study of 119 patients randomized to isotonic sodium
diagnosis with the characteristic features described above. bicarbonate (CIN incidence 1.7%) versus normal saline (CIN
incidence 13.6%) (41). Several subsequent trials have demon-
Prevention strated mixed results. The REMEDIAL trial showed a signifi-
The treatment of CIN is essentially supportive care for acute cant reduction in the incidence of CIN in 326 patients with
renal insufficiency, with attention paid to electrolyte and vol- mild to moderate baseline renal insufficiency (serum creatinine
ume abnormalities. The most effective treatment for CIN is > 2.0 mg/dL) treated with isotonic sodium bicarbonate þ
[14/6/010/13:1:45] [28–37]
acetylcysteine (1.9%) versus normal saline þ acetylcysteine randomized trial showed a marked reduction in CIN in 83
(9.9%) versus normal saline þ acetylcysteine þ vitamin C patients randomized to saline þ 600 mg twice daily acetylcys-
(10.3%) (42). A subsequent trial of 356 high-risk patients, how- teine versus saline alone (2% vs. 21%) (57). Further studies
ever, revealed no benefit of isotonic sodium bicarbonate com- suggest that 1200 mg twice daily may be superior to 600 mg
pared with isotonic saline alone (43). Two recent meta-analyses twice daily, with a reported CIN incidence of 8% versus 15%
of sodium bicarbonate studies concluded that it probably does (58). Some trials have shown equal benefit of IV and oral
reduce the incidence of CIN compared with saline alone, but acetylcysteine, but there is a risk of anaphylaxis with the IV
the results may be exaggerated by publication bias (44,45). form (58,59). Since the promising original trial, several random-
Nevertheless, routine prophylactic use of sodium bicarbonate ized prospective trials have shown no benefit for acetylcysteine
for prevention of CIN remains controversial. compared with hydration alone (60–62). Meta-analysis have
Type of contrast Some LOCM have been shown to demonstrated trends toward overall benefit, but the trials are
reduce the incidence of CIN in patients with preexisting mild heterogenous and the results are inconsistent (63). Given that
to moderate renal insufficiency when compared with HOCM the agent is well-tolerated and may have benefit, the use of
(5,7,16,46). A randomized trial of more than 1100 patients acetylcysteine for prevention of CIN in patients with chronic
referred for coronary angiography compared the LOCM renal insufficiency is routine at most centers.
iohexol with the HOCM diatrizoate (16). In patients with base- Inhibition of renal artery vasoconstriction Fenoldopam
line serum creatinine > 1.4 mg/dL, the incidence of CIN is a selective D-1 receptor agonist. In theory, it could improve
(defined as a rise in creatinine of 1.0 mg/dL) was significantly renal artery perfusion and attenuate the deleterious effects of
lower with iohexol compared with diatrizoate (7.2% vs. 15.8%). contrast media on the kidney. However, systemic infusions of
This difference was even more profound in diabetic patients fenodopam þ saline have failed to reduce CIN compared with
(11.8% vs. 27%). saline alone in two controlled studies (64,65). Intrarenal fenol-
The nonionic, IOCM iodixanol, has been compared with dopam using a bifurcated renal artery infusion catheter may be
various types of LOCM in a number of randomized trials. Some more effective than IV fenoldopam in improving GFR and
randomized trials have shown a reduced incidence of CIN with reducing the incidence of systemic hypotension and is cur-
iodixanol, while others have not (9,11,12,47). A meta-analysis of rently undergoing evaluation in randomised controlled trials
16 trials comparing iodixanol with LOCM concluded that (66). Another renal artery vasodilator, dopamine, has likewise
iodixanol was less nephrotoxic, but many of the trials were showed inconsistent promise in reducing CIN (67–69). Theo-
not designed with CIN as the primary endpoint, and thus the phylline, an adenosine A-1 receptor antagonist has had mixed
analysis may have suffered from ascertainment bias (10). In results in randomized trials (70,71). One randomized trial
addition, the difference was driven primarily by a large trial showed a significantly-reduced rise in creatinine of three dif-
which compared iodixanol with the ionic LOCM ioxaglate, ferent doses of prostaglandin E1 compared with placebo (72).
which is generally considered to be inferior to other, nonionic None of these measures have an evidence base to support their
LOCM. A more recent meta-analysis of 17 trials specifically routine clinical use.
showed no difference CIN between iodixanol and a pool of Other agents After an initial small randomized trial of
various LOCM (48). At this point, it is not clear that iodixanol is patients undergoing cardiac catheterization showed a reduced
superior to all nonionic LOCM, so large prospective random- incidence in CIN in patients who received oral vitamin C (9%)
ized trials are needed. versus placebo (20%), a larger trial showed no difference in
In the highest risk patients, there is some experience with saline þ acetylcysteine versus saline þ acetylcysteine þ vitamin
using gadolinium-containing contrast agents as a substitute for C (42). Recently, the cellular anti-ischemic agent trimetazidine
iodinated contrast (49–51). Gadolinium has rarely been associ- reduced CIN in a small randomized trial (73).
ated with nephrotoxicity, but it should be used with caution in Hemodialysis and hemofiltration Prophylactic hemo-
patients with moderate to severe renal failure, given the dialysis does not appear to diminish the incidence of CIN in
increased risk of nephrogenic systemic fibrosis. CO2 has also patients with preexisting renal insufficiency (74). A meta-anal-
been used in the peripheral vasculature, but it must be limited ysis of 412 patients from six studies with a creatinine ranging
to below the diaphragm because of the potential risk of cerebral from 2.5 to 4.0 mg/dL showed no benefit and there was a
embolization, and digital subtraction angiography must be suggestion of harm (75). Prophylactic continuous venovenous
used (52,53). hemofiltration (CVVH) in moderate to severe renal insuffi-
Volume of contrast Volume of contrast administered is ciency (mean creatinine 3.0 mg/dL) results in a lower creati-
crucial to the development of CIN, and minimization of con- nine and a lower mortality compared with standard of care, but
trast volume is of utmost importance in its prevention (54,55). the outcomes may have resulted from the fact that CVVH
The mean volume of contrast administered during cardiac removes creatinine and those patients in the CVVH arm were
catheterization is 130 mL for diagnostic procedures and 191 all treated in the intensive care unit; their greater intensity of
mL for interventional procedures (56). In patients with chronic care compared with the control arm may have explained their
renal insufficiency or other conditions which put them at risk improved short- and long-term survival (76). A recent trial in
for CIN, techniques such as use of biplane cameras, limiting patients with severe chronic kidney disease (serum creatinine
contrast puffs, minimizing cine runs, or staging interventional 4.9 mg/dL) referred for cardiac catheterization found a
procedures until at least 72 hours after a diagnostic procedure reduced incidence of long-term need for hemodialysis in
can be useful. those who received immediate hemodialysis compared with
the control group (0% vs. 13%, respectively) (77). There may be
Potential Prophylactic Therapy a role for prophylactic hemodialysis in a subpopulation of
Antioxidant therapy Given the possibility of contrast- patients who are at the highest risk of CIN but more studies
mediated oxidative kidney injury in CIN, prophylactic use of are warranted before routine use of this strategy can be
the antioxidant, acetylcysteine has been evaluated. An early recommended.
[14/6/010/13:1:45] [28–37]
CONTRAST AGENTS 33
HYPERSENSITIVITY REACTIONS been reported with only 2 mL of contrast (94). Symptoms

Unlike chemotoxic contrast reactions, hypersensitivity reac- typically begin within minutes up to one hour after contrast
tions are independent of the rate or volume of contrast infusion. infusion. Reactions which occur within seconds after contrast
Immediate hypersensitivity reactions typically occur within exposure may initially seem mild but tend to progress rapidly.
seconds to an hour after contrast exposure. Delayed hypersen- Typical manifestations include pruritus and urticaria. More
sitivity reactions can occur between one hour and one week, severe reactions may include angioedema, laryngospasm, bron-
but this is much less common. Hypersensitivity reactions typ- chospasm, loss of consciousness, and even death. Danger signs
ically present with some combination of pruritus, urticaria, are rapid progression of symptoms, respiratory distress, stri-
rash, angioedema, laryngospasm, bronchospasm, hypotension, dor, hypotension, arrhythmia, or chest pain.
syncope, shock, or even death. Most such reactions are mild, It is important to distinguish an anaphylactoid reaction
but they can (rarely) be fatal. from other complications of contrast injection, such as vasovagal
reactions or cardiogenic pulmonary edema. Like an anaphylac-
toid reaction, a vasovagal reaction can result in acute hypoten-
Epidemiology sion, but typically involves bradycardia and nausea, and
The incidence of hypersensitivity reactions has been as high as patients will not have pruritus, urticaria, angioedema, stridor,
13% in past analyses, but is much lower with newer agents or wheezing. Rapid volume expansion after contrast bolus can
(78,79). Overall, the incidence of mild reactions is approxi- cause respiratory distress and wheezing from cardiogenic pul-
mately 9%, and severe reactions 0.2% to 1.6% (80,81). The monary edema, but this syndrome likewise will not be associ-
mortality is <1/100,000 patients (82). It is seven times higher ated with puritus, urticaria, or angioedema.
in the elderly, in part related to comorbidities (83). Lab testing for anaphylactoid reactions is not routinely
done, but there are two tests that may help indicate that an
anaphylactoid reaction has occurred: serum tryptase, and
Pathophysiology
24-hour urine histamine. Tryptase is a mast cell proteinase
The pathophysiology of hypersensitivity reactions is poorly
which is released in large quantities in the setting of massive
understood. The clinical manifestations may be similar to that
mast cell activation and degranulation such as anaphylaxis,
of anaphylaxis, but unlike anaphylaxis, the vast majority of
anaphylactoid reactions, and systemic mastocytosis (84,95). It
contrast hypersensitivity reactions are not IgE-mediated (84,85).
has a 90-minute half-life and is only detectable for a couple of
The most likely etiology is an anaphylactoid reaction involving
hours after an anaphylactoid reaction. Histamine has an even
direct mast cell activation and massive degranulation of hista-
shorter half-life. Elevated histamine levels can be detected on a
mine. Activation of the coagulation cascade and kinin, release
24-hour urine collection, which is begun soon after an anaphy-
of serotonin, and inhibition of platelets may also play a role
lactoid reaction.
(86–88).
Routine skin testing has not proven to be beneficial in
predicting the recurrence of a hypersensitivity reaction to con-
Risk Factors trast media, although there may be a very small subset of
The type of contrast media used impacts the likelihood of a patients with true IgE-mediated anaphylaxis who can be
hypersensitivity reaction. Ionic HOCM cause mild to moderate detected with skin testing (85).
reactions in as many as 12% of patients, compared with less
than 3% with nonionic LOCM (89). Severe reactions occur in
0.22% to 0.04% with ionic HOCM, compared with 0.04% to Treatment
0.004% with nonionic LOCM. The incidence of hypersensitivity The initial step in the treatment of a presumed anaphylactoid
reactions may be even lower with the nonionic IOCM iodixanol contrast reaction is to stop injecting contrast medium (Table 4.3).
(78,79). Mild reactions, such as simple pruritus, may be treated with
The most important risk factor for the development of an antihistamines and observation. However, it should be noted
anaphylactoid reaction is a prior anaphylactoid reaction to that mild reactions can quickly progress, and the clinician
contrast media, with repeat unpremedicated reaction rates as should anticipate and respond to a severe reaction when
high as 50% in some observational studies (89,90). Other risk needed.
factors include asthma or an allergic disease history (89). The first and most important active treatment of a severe
It has been proposed that shellfish allergy confers a anaphylactoid reaction is epinephrine. Other therapies take
higher risk of contrast allergy because shellfish contain iodine time to work, and may not be effective in reversing the under-
(91,92). This is likely not the case. In fact, the allergen in lying reaction. Aqueous intramuscular epinephrine 0.3 to
shellfish is the protein tropomysin, not iodine. However, atopic 0.5 mg in the anterolateral thigh can be given. Alternatively, a
individuals are commonly allergic to seafood, and atopic 0.1 mg vial of epinephrine from the crash cart can be diluted in
individuals tend to be at higher risk of a contrast reaction. In 10 mL of saline and given 1 mL (10 mcg) at a time intra-
addition, patients who react to topical iodine are not at venously. If symptoms persist despite epinephrine boluses,
increased risk of an anaphylactoid reaction to contrast media. then an IV drip at 2 to 10 mcg/min can be infused. Patients
It has been proposed that certain medications, such as on b-blockers may not respond adequately to epinephrine. In
b-blockers and NSAIDs, may increase the risk of a contrast this setting, glucagons 1 to 2 mg IV can be given over five
reaction, but this has not been definitively established (90,93). minutes, followed by an infusion of 5 to 15 mcg/min.
In addition to oxygen, albuterol nebulizer may be given
to relieve bronchospasm. If there is evidence of impending
Clinical Features airway obstruction from angioedema or stridor, immediate
Hypersensitivity reactions are not related to the dose or the endotracheal intubation should take place. Delay may result
injection rate of contrast medium. In fact, fatal reactions have in complete obstruction requiring cricothyroidotomy.
[14/6/010/13:1:45] [28–37]
Table 4.3 Prophylaxis and Treatment of Contrast Hypersensitivity Reactions

Prophylaxis in setting of known Prednisone 50 mg PO, 13, 7, and 1 hr before procedure
contrast allergy
Or hydrocortizone (for emergent 200 mg IV, prior to and every 4 hr during an
procedures only) emergent procedure
Diphenhydramine 50 mg IV, 1 hr before procedure
Ranitidine 50 mg IV or PO, 1 hr before procedure
Use low-osmolar or iso-osmolar
contrast media
Treatment in setting of mild Stop contrast injection 50 mg IV
contrast reaction Diphenhydramine
Ranitidine 50 mg IV or PO
Observe for progression to severe
reaction
Treatment in setting of severe Stop contrast injection
contrast reaction Protect airway Immediate intubation if stridor or angioedema
with airway obstruction
Epinephrine 10 mcg IV followed by 2–10 mcg/min IV if
symptoms persist or 0.3–0.5 mg IM in
anterolateral thigh
Oxygen 6 L up to 100% facemask as needed
Normal saline 1 L IV bolus, repeat as needed
Albuterol Nebulizer
Glucagon (if on b-blocker and not 1–2 mg IV over 5 min, followed by an infusion
responding to epinephrine) of 5–15 mcg/min
Methyl prednisone 125 mg IV (to prevent recurrent reaction
several hours later)
Diphenhydramine 50 mg IV
Ranitidine 50 mg IV or PO
Adjunctive antihistamine therapy with IV diphenhydr- Gadolinium is even less likely to cause a repeat contrast
amine 50 mg, an H1 blocker and ranitidine 50 mg, an H2 reaction, with the incidence ranging from 1/100,000 to
blocker, may improve symptoms and help prevent short-term 1/500,000 cases and essentially confined to rare case reports
recurrence. Glucocorticoids have little immediate benefit, but (99). In addition, CO2 can be used for peripheral interventions,
methylprednisolone 125 mg IV can be given to prevent a but must be confined to injections below the diaphragm given
recurrent reaction several hours later. the risk of cerebral embolization.
There are no routine empiric medications which are
recommended for prevention of first-time reactions. If not
Prevention already a routine practice, one may consider using LOCM in
As indicated earlier, the most important risk factor for a con-
patients with asthma or with any history of other serious
trast reaction is a history of previous reactions. Premedication
allergic reactions. Most prevention is directed at minimizing
is crucial to preventing a repeat reaction. The combination of
recurrent reactions.
prednisone, diphenhydramine, and LOCM lowers the risk of a
recurrent reaction to less than 1%. The exact dosing of gluco-
corticoids is not known, but glucocorticoids need to be taken
SUMMARY
Contrast agents allow visualization of the cardiac anatomy.
hours in advance to have the maximum effect. A validated
Many different formulations have been introduced in an attempt
protocol is prednisone 50 mg by mouth at 13 hours, 7 hours,
to minimize adverse side. In general, LOCM and IOCM are
and 1 hour prior to the procedure (96,97). In addition, diphen-
equally as effective as and are associated with less toxicity than
hydramine 50 mg should be given one hour prior to the
the HOCM. It is not entirely clear whether IOCM are safer than
procedure.
LOCM in the prevention of CIN. Many evidence-based methods
In the case of an emergency procedure, a rapid premed-
have been developed to prevent anaphylactoid contrast reac-
ication protocol may include hydrocortisone 200 mg IV once
tions and CIN, particularly in patients considered to be at higher
and then every four hours until the procedure is complete,
risk. These methods should be used on a routine basis in the
diphenhydramine 50 mg IV or orally, and LOCM.
cardiac catheterization laboratory.
There is no clear benefit or harm with prophylactic H2
blockers (90,98). Ephedrine has been shown to be useful, but
this drug is relatively contraindicated in patients with coronary REFERENCES
1. Baim DS, ed. Grossman’s Cardiac Catheterization, Angiography,
artery disease and may promote coronary vasospasm.
and Intervention. 7th ed. Philadelphia, PA: Lippincott Williams
Use of LOCM reduces the chance of recurrence compared & Wilkins, 2006.
with HOCM. Use of the IOCM iodixanol, may reduce this risk 2. Zukerman LS, Friehling TD, Wolf NM, et al. Effect of calcium-
even further. The incidence of a repeat contrast reaction using binding additives on ventricular fibrillation and repolarization
iodixanol was 0.7%, compared with 2% when using the ionic, changes during coronary angiography. J Am Coll Cardiol 1987;
LOCM ioxaglate (4). 10:12491253.
[14/6/010/13:1:45] [28–37]
CONTRAST AGENTS 35
3. Piao ZE, Murdock DK, Hwang MH, et al. Hemodynamic abnor- 22. Rudnick M, Feldman H. Contrast-induced nephropathy: what
malities during coronary angiography: comparison of Hypaque- are the true clinical consequences? Clin J Am Soc Nephrol 2008;
76, Hexabrix, and Omnipaque-350. Cathet Cardiovasc Diagn 3:263–272.
1989; 16:149154. 23. Heyman SN, Reichman J, Brezis M. Pathophysiology of radio-
4. Bertrand ME, Esplugas E, Piessens J, et al. Influence of a non- contrast nephropathy: a role for medullary hypoxia. Invest
ionic, iso-osmolar contrast medium (iodixanol) versus an ionic, Radiol 1999; 34:685–691.
low-osmolar contrast medium (ioxaglate) on major adverse car- 24. Persson PB, Hansell P, Liss P. Pathophysiology of contrast
diac events in patients undergoing percutaneous transluminal medium-induced nephropathy. Kidney Int 2005; 68:14–22.
coronary angioplasty: a multicenter, randomized, double-blind 25. Klause N, Arendt T, Lins M, et al. Hypoxic renal tissue damage
study. Visipaque in percutaneous transluminal coronary angio- by endothelin-mediated arterial vasoconstriction during radio-
plasty [VIP] trial investigators. Circulation 2000; 101:131136. angiography in man. Adv Exp Med Biol 1998; 454:225–234.
5. Moore RD, Steinberg EP, Powe NR, et al. Nephrotoxicity of high- 26. Cantley LG, Spokes K, Clark B, et al. Role of endothelin and
osmolality versus low-osmolality contrast media: randomized prostaglandins in radiocontrast-induced renal artery constric-
clinical trial. Radiology 1992; 182:649655. tion. Kidney Int 1993; 44:1217–1223.
6. Barrett BJ, Parfrey PS, Vavasour HM, et al. Contrast nephropathy 27. Katholi RE, Taylor GJ, McCann WP, et al. Nephrotoxicity from
in patients with impaired renal function: high versus low osmo- contrast media: attenuation with theophylline. Radiology 1995;
lar media. Kidney Int 1992; 41:1274–1279. 195:17–22.
7. Barrett BJ, Carlisle EJ. Metaanalysis of the relative nephrotoxicity 28. Pflueger A, Larson TS, Nath KA, et al. Role of adenosine in
of high- and low-osmolality iodinated contrast media. Radiology contrast media-induced acute renal failure in diabetes mellitus.
1993; 188:171–178. Mayo Clin Proc 2000; 75:1275–1283.
8. Sutton AG, Ashton VJ, Campbell PG, et al. A randomized pro- 29. Elkayam U, Mehra A, Cohen G, et al. Renal circulatory effects of
spective trial of ioxaglate 320 (Hexabrix) vs. iodixanol 320 adenosine in patients with chronic heart failure. J Am Coll
(Visipaque) in patients undergoing percutaneous coronary inter- Cardiol 1998; 32:211–215.
vention. Catheter Cardiovasc Interv 2002; 57:346–352. 30. Gussenhoven MJ, Ravensbergen J, van Bockel JH, et al. Renal
9. Jo SH, Youn TJ, Koo BK, et al. Renal toxicity evaluation and dysfunction after angiography; a risk factor analysis in patients
comparison between Visipaque (iodixanol) and Hexabrix (ioxa- with peripheral vascular disease. J Cardiovasc Surg (Torino)
glate) in patients with renal insufficiency undergoing coronary 1991; 32:81–86.
angiography: the RECOVER study: a randomized controlled 31. McCullough PA, Wolyn R, Rocher LL, et al. Acute renal failure
trial. J Am Coll Cardiol 2006; 48:924–930. after coronary intervention: incidence, risk factors, and relation-
10. McCullough PA, Bertrand ME, Brinker JA, et al. A meta-analysis ship to mortality. Am J Med 1997; 103:368–375.
of the renal safety of isosmolar iodixanol compared with low- 32. Mehran R, Aymong ED, Nikolsky E, et al. A simple risk score for
osmolar contrast media. J Am Coll Cardiol 2006; 48:692–699. prediction of contrast-induced nephropathy after percutaneous
11. Solomon RJ, Natarajan MK, Doucet S, et al. Cardiac angiography coronary intervention: development and initial validation. J Am
in renally impaired patients (CARE) study: a randomized dou- Coll Cardiol 2004; 44:1393–1399.
ble-blind trial of contrast-induced nephropathy in patients with 33. Nikolsky E, Mehran R, Lasic Z, et al. Low hematocrit predicts
chronic kidney disease. Circulation 2007; 115:3189–3196. contrast-induced nephropathy after percutaneous coronary inter-
12. Rudnick MR, Davidson C, Laskey W, et al. Nephrotoxicity of ventions. Kidney Int 2005; 67:706–713.
iodixanol versus ioversol in patients with chronic kidney disease: 34. Cirit M, Toprak O, Yesil M, et al. Angiotensin-converting enzyme
the Visipaque angiography/interventions with laboratory out- inhibitors as a risk factor for contrast-induced nephropathy.
comes in renal insufficiency (VALOR) trial. Am Heart J 2008; Nephron Clin Pract 2006; 104:c20c27.
156:776–782. 35. Gupta RK, Kapoor A, Tewari S, et al. Captopril for prevention of
13. Liss P, Persson PB, Hansell P, et al. Renal failure in 57925 patients contrast-induced nephropathy in diabetic patients: a randomised
undergoing coronary procedures using iso-osmolar or low- study. Indian Heart J 1999; 51:521–526.
osmolar contrast media. Kidney Int 2006; 70:1811–1817. 36. Rudnick MR, Berns JS, Cohen RM, et al. Nephrotoxic risks of
14. Parfrey PS, Griffiths SM, Barrett BJ, et al. Contrast material- renal angiography: contrast media-associated nephrotoxicity
induced renal failure in patients with diabetes mellitus, renal and atheroembolism–a critical review. Am J Kidney Dis 1994;
insufficiency, or both. A prospective controlled study. N Engl J 24:713–727.
Med 1989; 320:143–149. 37. Solomon R, Werner C, Mann D, et al. Effects of saline, mannitol,
15. Rich MW, Crecelius CA. Incidence, risk factors, and clinical and furosemide to prevent acute decreases in renal function
course of acute renal insufficiency after cardiac catheterization induced by radiocontrast agents. N Engl J Med 1994; 331:1416–1420.
in patients 70 years of age or older. A prospective study. Arch 38. Mueller C, Buerkle G, Buettner HJ, et al. Prevention of contrast
Intern Med 1990; 150:1237–1242. media-associated nephropathy: randomized comparison of 2
16. Rudnick MR, Goldfarb S, Wexler L, et al. Nephrotoxicity of ionic hydration regimens in 1620 patients undergoing coronary angio-
and nonionic contrast media in 1196 patients: a randomized trial. plasty. Arch Intern Med 2002; 162:329–336.
The Iohexol Cooperative Study. Kidney Int 1995; 47:254–261. 39. Taylor AJ, Hotchkiss D, Morse RW, et al. PREPARED: Prepara-
17. Schwab SJ, Hlatky MA, Pieper KS, et al. Contrast nephrotoxicity: tion for angiography in renal dysfunction: a randomized trial of
a randomized controlled trial of a nonionic and an ionic radio- inpatient vs outpatient hydration protocols for cardiac catheter-
graphic contrast agent. N Engl J Med 1989; 320:149–153. ization in mild-to-moderate renal dysfunction. Chest 1998;
18. Manske CL, Sprafka JM, Strony JT, et al. Contrast nephropathy in 114:1570–1574.
azotemic diabetic patients undergoing coronary angiography. 40. Trivedi HS, Moore H, Nasr S, et al. A randomized prospective
Am J Med 1990; 89:615–620. trial to assess the role of saline hydration on the development of
19. Rihal CS, Textor SC, Grill DE, et al. Incidence and prognostic contrast nephrotoxicity. Nephron Clin Pract 2003; 93:C29C34.
importance of acute renal failure after percutaneous coronary 41. Merten GJ, Burgess WP, Gray LV, et al. Prevention of contrast-
intervention. Circulation 2002; 105:2259–2264. induced nephropathy with sodium bicarbonate: a randomized
20. Gupta R, Gurm HS, Bhatt DL, et al. Renal failure after percuta- controlled trial. JAMA 2004; 291:2328–2334.
neous coronary intervention is associated with high mortality. 42. Briguori C, Airoldi F, D’Andrea D, et al. Renal Insufficiency
Catheter Cardiovasc Interv 2005; 64:442–448. Following Contrast Media Administration Trial (REMEDIAL): a
21. Levy EM, Viscoli CM, Horwitz RI. The effect of acute renal failure randomized comparison of 3 preventive strategies. Circulation
on mortality. A cohort analysis. JAMA 1996; 275:1489–1494. 2007; 115:1211–1217.
[14/6/010/13:1:45] [28–37]
43. Brar SS, Shen AY, Jorgensen MB, et al. Sodium bicarbonate vs 63. Nallamothu BK, Shojania KG, Saint S, et al. Is acetylcysteine
sodium chloride for the prevention of contrast medium-induced effective in preventing contrast-related nephropathy? A meta-
nephropathy in patients undergoing coronary angiography: a analysis. Am J Med 2004; 117:938–947.
randomized trial. JAMA 2008; 300:1038–1046. 64. Allaqaband S, Tumuluri R, Malik AM, et al. Prospective random-
44. Hogan SE, L’Allier P, Chetcuti S, et al. Current role of sodium ized study of N-acetylcysteine, fenoldopam, and saline for preven-
bicarbonate-based preprocedural hydration for the prevention of tion of radiocontrast-induced nephropathy. Catheter Cardiovasc
contrast-induced acute kidney injury: a meta-analysis. Am Heart Interv 2002; 57:279–283.
J 2008; 156:414–421. 65. Stone GW, McCullough PA, Tumlin JA, et al. Fenoldopam
45. Meier P, Ko DT, Tamura A, et al. Sodium bicarbonate-based mesylate for the prevention of contrast-induced nephropathy: a
hydration prevents contrast-induced nephropathy: a meta- randomized controlled trial. JAMA 2003; 290:2284–2291.
analysis. BMC Med 2009; 7:23. 66. Teirstein PS, Price MJ, Mathur VS, et al. Differential effects
46. Lautin EM, Freeman NJ, Schoenfeld AH, et al. Radiocontrast- between intravenous and targeted renal delivery of fenoldopam
associated renal dysfunction: a comparison of lower-osmolality on renal function and blood pressure in patients undergoing
and conventional high-osmolality contrast media. AJR Am J cardiac catheterization. Am J Cardiol 2006; 97:1076–1081.
Roentgenol 1991; 157:59–65. 67. Kapoor A, Sinha N, Sharma RK, et al. Use of dopamine in
47. Aspelin P, Aubry P, Fransson SG, et al. Nephrotoxic effects in prevention of contrast induced acute renal failure–a randomised
high-risk patients undergoing angiography. N Engl J Med 2003; study. Int J Cardiol 1996; 53:233–236.
348:491–499. 68. Gare M, Haviv YS, Ben-Yehuda A, et al. The renal effect of low-
48. Reed M, Meier P, Tamhane UU, et al. The relative renal safety dose dopamine in high-risk patients undergoing coronary
of iodixanol compared with low-osmolar contrast media: a angiography. J Am Coll Cardiol 1999; 34:1682–1688.
meta-analysis of randomized controlled trials. JACC Cardiovasc 69. Hans SS, Hans BA, Dhillon R, et al. Effect of dopamine on renal
Interv 2009; 2:645–654. function after arteriography in patients with pre-existing renal
49. Kaufman JA, Geller SC, Bazari H, et al. Gadolinium-based insufficiency. Am Surg 1998; 64:432–436.
contrast agents as an alternative at vena cavography in patients 70. Huber W, Ilgmann K, Page M, et al. Effect of theophylline on
with renal insufficiency–early experience. Radiology 1999; 212: contrast material-nephropathy in patients with chronic renal
280–284. insufficiency: controlled, randomized, double-blinded study.
50. Kaufman JA, Hu S, Geller SC, et al. Selective angiography of the Radiology 2002; 223:772–779.
common carotid artery with gadopentetate dimeglumine in a 71. Erley CM, Duda SH, Rehfuss D, et al. Prevention of radiocon-
patient with renal insufficiency. AJR Am J Roentgenol 1999; trast-media-induced nephropathy in patients with pre-existing
172:1613–1614. renal insufficiency by hydration in combination with the adeno-
51. Boyden TF, Gurm HS. Does gadolinium-based angiography sine antagonist theophylline. Nephrol Dial Transplant 1999;
protect against contrast-induced nephropathy?: a systematic 14:1146–1149.
review of the literature. Catheter Cardiovasc Interv 2008; 72. Sketch MH Jr, Whelton A, Schollmayer E, et al. Prevention of
71:687–693. contrast media-induced renal dysfunction with prostaglandin E1:
52. Kerns SR, Hawkins IF Jr. Carbon dioxide digital subtraction a randomized, double-blind, placebo-controlled study. Am J
angiography: expanding applications and technical evolution. Ther 2001; 8:155–162.
AJR Am J Roentgenol 1995; 164:735–741. 73. Onbasili AO, Yeniceriglu Y, Agaoglu P, et al. Trimetazidine in
53. Kerns SR, Hawkins IF Jr, Sabatelli FW. Current status of carbon the prevention of contrast-induced nephropathy after coronary
dioxide angiography. Radiol Clin North Am 1995; 33:15–29. procedures. Heart 2007; 93:698–702.
54. Kahn JK, Rutherford BD, McConahay DR, et al. High-dose con- 74. Vogt B, Ferrari P, Schonholzer C, et al. Prophylactic hemodialysis
trast agent administration during complex coronary angioplasty. after radiocontrast media in patients with renal insufficiency is
Am Heart J 1990; 120:533–536. potentially harmful. Am J Med 2001; 111:692–698.
55. Rosovsky MA, Rusinek H, Berenstein A, et al. High-dose admin- 75. Cruz DN, Perazella MA, Bellomo R, et al. Extracorporeal blood
istration of nonionic contrast media: a retrospective review. purification therapies for prevention of radiocontrast-induced
Radiology 1996; 200:119–122. nephropathy: a systematic review. Am J Kidney Dis 2006;
56. Noto TJ Jr, Johnson LW, Krone R, et al. Cardiac catheterization 48:361–371.
1990: a report of the Registry of the Society for Cardiac Angiog- 76. Marenzi G, Marana I, Lauri G, et al. The prevention of radio-
raphy and Interventions (SCA&I). Cathet Cardiovasc Diagn 1991; contrast-agent-induced nephropathy by hemofiltration. N Engl J
24:75–83. Med 2003; 349:1333–1340.
57. Tepel M, van der Giet M, Schwarzfeld C, et al. Prevention of 77. Lee PT, Chou KJ, Liu CP, et al. Renal protection for coronary
radiographic-contrast-agent-induced reductions in renal function angiography in advanced renal failure patients by prophylactic
by acetylcysteine. N Engl J Med 2000; 343:180–184. hemodialysis. A randomized controlled trial. J Am Coll Cardiol
58. Marenzi G, Assanelli E, Marana I, et al. N-acetylcysteine and 2007; 50:1015–1020.
contrast-induced nephropathy in primary angioplasty. N Engl J 78. Sutton AG, Finn P, Grech ED, et al. Early and late reactions after
Med 2006; 354:2773–2782. the use of iopamidol 340, ioxaglate 320, and iodixanol 320 in
59. Baker CS, Wragg A, Kumar S, et al. A rapid protocol for the cardiac catheterization. Am Heart J 2001; 141:677–683.
prevention of contrast-induced renal dysfunction: the RAPPID 79. Sutton AG, Finn P, Campbell PG, et al. Early and late reactions
study. J Am Coll Cardiol 2003; 41:2114–2118. following the use of iopamidol 340, iomeprol 350 and iodixanol
60. Azmus AD, Gottschall C, Manica A, et al. Effectiveness of 320 in cardiac catheterization. J Invasive Cardiol 2003; 15:133–138.
acetylcysteine in prevention of contrast nephropathy. J Invasive 80. Thomsen HS, Morcos SK. Management of acute adverse reac-
Cardiol 2005; 17:80–84. tions to contrast media. Eur Radiol 2004; 14:476–481.
61. Kay J, Chow WH, Chan TM, et al. Acetylcysteine for prevention 81. Tramer MR, von Elm E, Loubeyre P, et al. Pharmacological
of acute deterioration of renal function following elective coro- prevention of serious anaphylactic reactions due to iodinated
nary angiography and intervention: a randomized controlled contrast media: systematic review. BMJ 2006; 333:675.
trial. JAMA 2003; 289:553–558. 82. Caro JJ, Trindade E, McGregor M. The risks of death and of
62. Webb JG, Pate GE, Humphries KH, et al. A randomized con- severe nonfatal reactions with high- vs low-osmolality contrast
trolled trial of intravenous N-acetylcysteine for the prevention of media: a meta-analysis. AJR Am J Roentgenol 1991; 156:825–832.
contrast-induced nephropathy after cardiac catheterization: lack 83. Cashman JD, McCredie J, Henry DA. Intravenous contrast media:
of effect. Am Heart J 2004; 148:422–429. use and associated mortality. Med J Aust 1991; 155:618–623.
[14/6/010/13:1:45] [28–37]
CONTRAST AGENTS 37
84. Laroche D, Aimone-Gastin I, Dubois F, et al. Mechanisms of 92. Beaty AD, Lieberman PL, Slavin RG. Seafood allergy and radio-
severe, immediate reactions to iodinated contrast material. Radi- contrast media: are physicians propagating a myth? Am J Med
ology 1998; 209:183–190. 2008; 121:158.e1158.e4.
85. Trcka J, Schmidt C, Seitz CS, et al. Anaphylaxis to iodinated 93. Lang DM, Alpern MB, Visintainer PF, et al. Elevated risk of
contrast material: nonallergic hypersensitivity or IgE-mediated anaphylactoid reaction from radiographic contrast media is
allergy? AJR Am J Roentgenol 2008; 190:666–670. associated with both beta-blocker exposure and cardiovascular
86. Simon RA, Schatz M, Stevenson DD, et al. Radiographic contrast disorders. Arch Intern Med 1993; 153:2033–2040.
media infusions. Measurement of histamine, complement, and 94. Fischer HW, Doust VL. An evaluation of pretesting in the prob-
fibrin split products and correlation with clinical parameters. J lem of serious and fatal reactions to excretory urography. Radi-
Allergy Clin Immunol 1979; 63:281–288. ology 1972; 103:497–501.
87. Ring J, Simon RA, Arroyave CM. Increased in vitro histamine 95. Brockow K, Vieluf D, Puschel K, et al. Increased postmortem
release by radiographic contrast media in patients with history of serum mast cell tryptase in a fatal anaphylactoid reaction to
incompatibility. Clin Exp Immunol 1978; 34:302–309. nonionic radiocontrast medium. J Allergy Clin Immunol 1999;
88. Zir LM, Carvalho AC, Hawthorne JW, et al. Effect of contrast 104:237–238.
agents on platelet aggregation and 14C-serotonin release. N Engl 96. Greenberger PA, Patterson R. The prevention of immediate
J Med 1974; 291:134–135. generalized reactions to radiocontrast media in high-risk
89. Katayama H, Yamaguchi K, Kozuka T, et al. Adverse reactions to patients. J Allergy Clin Immunol 1991; 87:867–872.
ionic and nonionic contrast media. A report from the Japanese 97. Greenberger PA, Patterson R. Adverse reactions to radiocontrast
Committee on the Safety of Contrast Media. Radiology 1990; media. Prog Cardiovasc Dis 1988; 31:239–248.
175:621–628. 98. Marshall GD Jr, Lieberman PL. Comparison of three pretreat-
90. Greenberger PA, Patterson R, Tapio CM. Prophylaxis against ment protocols to prevent anaphylactoid reactions to radiocon-
repeated radiocontrast media reactions in 857 cases. Adverse trast media. Ann Allergy 1991; 67:70–74.
experience with cimetidine and safety of beta-adrenergic antag- 99. Dillman JR, Ellis JH, Cohan RH, et al. Allergic-like breakthrough
onists. Arch Intern Med 1985; 145:2197–2200. reactions to gadolinium contrast agents after corticosteroid and
91. Huang SW. Seafood and iodine: an analysis of a medical myth. antihistamine premedication. AJR Am J Roentgenol 2008;
Allergy Asthma Proc 2005; 26:468–469. 190:187–190.
[14/6/010/12:55:6] [38–46]
Patient selection, preparation, risks, and informed consent

David O. Williams and J. Dawn Abbott
INTRODUCTION perform the necessary procedural components. For patients

Cardiac catheterization is one of the most common in-hospital that subsequently require interventional procedures including
procedures performed in the United States with an annual those detailed in the following chapters of this book, clinical
volume of over one million and the number is expected to guidelines from the ACC/AHA exist for many diseases and
increase as the population ages. The procedures performed in procedures and can serve as a framework for making clinical
the cardiac catheterization laboratory have expanded over time decisions.
to include the diagnosis and treatment of a wide range of Cardiac catheterization may be performed as part of a
cardiac and vascular diseases. Independent of the type or diagnostic or therapeutic research study, or solely for research
complexity, all procedures involve the insertion of catheters purposes. In these circumstances, the protocol must be
into the circulatory system guided by a combination of fluo- approved by the local Institutional Review Board, the patient
roscopy, hemodynamic monitoring, and contrast media injec- informed of the investigative nature of the study, and informed
tion. Paramount to patient safety is an experienced and consent obtained.
competent catheterization laboratory team. The physician per-
forming the procedure in each case is responsible for determin-
ing the potential risks and benefits of a given procedure on the Procedural Contraindications
basis of both patient and technical factors. Proper patient As the field of interventional cardiology has evolved to rou-
selection and preparation can minimize the occurrence or tinely care for critically ill patients such as those with myocar-
severity of complications. The risks of angiography should be dial infarction (MI) and cardiogenic shock, the number of
understood by all members of the catheterization laboratory absolute contraindications to cardiac catheterization has
team so that complications can be anticipated and responded to decreased. The majority of procedures are still done on an
in a rapid and coordinated manner. An in-depth knowledge of elective or urgent, nonemergent basis and a number of relative
the risks of each procedure is also required to obtain appropri- contraindications to the procedure have been reported that can
ate informed consent from the patient. The preprocedural often be addressed prior to the procedure to improve safety
aspects of cardiac catheterization deserve special review and (Table 5.2). When possible, procedures should be delayed until
are the basis for this chapter. all relative contraindications are evaluated and treated.
The list of relative contraindications includes severe
contrast allergy, which is uncommon but potentially life-
PATIENT SELECTION threatening. Proper pretreatment, discussed below, can reduce
Procedural Indications the incidence and severity of recurrent reactions. Acute or
Before a decision to recommend cardiac catheterization is made chronic kidney disease may alter the risk/benefit ratio of car-
the indication for and alternatives to the procedure should be diac catheterization or may result in the requirement for
clear. The patient should know what clinical information is hemodialysis. Measures such as ensuring adequate volume
desired and how the findings will be used. An invasive study status before treatment to prevent contrast nephropathy, and
should be performed for diagnostic or prognostic indications consultation with a nephrologist may be required in such cases.
when incremental to a noninvasive evaluation or for therapeu- Uncontrolled ventricular irritability may increase the incidence
tic purposes. A thorough review of the patient’s medical his- of ventricular tachycardia or fibrillation. Electrolyte imbalances
tory, physical exam, electrocardiogram, and laboratory and should be corrected, diagnosis and treatment of potential drug
cardiac testing results is required to determine the appropri- toxicity should be sought, and medical therapy for ischemia
ateness of and type of procedure planned. Left heart catheter- optimized. Attention to control of hypertension can reduce the
ization is commonly performed for coronary artery disease and potential for ischemia, congestive heart failure, and bleeding
includes aortic and left ventricular pressures, left ventriculog- related to angiography. Patients with decompensated conges-
raphy, and coronary angiography. The practice guidelines for tive heart failure in the absence of acute myocardial ischemia or
coronary angiography are detailed in a report from the Amer- the need for hemodynamic support should be stabilized prior
ican College of Cardiology and American Heart Association to catheterization when possible. Abnormal coagulation or
(ACC/AHA) and are summarized in Table 5.1 (1). blood counts should be corrected when feasible to reduce
In many clinical scenarios such as valvular, myocardial, bleeding risk and ischemic complications. The presence of a
and congenital heart disease, one or more additional proce- febrile illness or bacteremia is an additional relative contra-
dures such as aortography, right heart catheterization, or indication. Several of the relative contraindications will be
oximetry are required for a complete diagnostic study. The determined and treatment optimized during patient prepara-
operator must, therefore, have the expertise to determine and tion for cardiac catheterization.
[14/6/010/12:55:6] [38–46]
PATIENT SELECTION, PREPARATION, RISKS, AND INFORMED CONSENT 39
Table 5.1 Indications for Coronary Angiography in Patients with Known or Suspected Coronary Artery Disease
Asymptomatic or stable angina Unstable coronary syndromes
l High-risk criteria on noninvasive testing l Unstable angina refractory to medical therapy
l Sudden cardiac death survivors l Unstable angina with intermediate to high short term risk
l Sustained monomorphic or polymorphic ventricular tachycardia l Unstable angina with subsequent abnormal noninvasive testing
l Angina on medical treatment l Suspected Prinzmetal variant angina
l Worsening abnormalities on serial noninvasive testing l Suspected stent thrombosis
l Patients that cannot be adequately risk stratified by other means l Patients with STEMI with the intent to perform primary or rescue
l Individuals whose occupation involves the safety of others who percutaneous coronary intervention
have an abnormal stress test or high-risk clinical features l Evaluation of non-STEMI
l Abnormal but not high-risk stress test in a patient with high l Cardiogenic shock
likelihood of disease l Post myocardial infarction with angina, congestive heart failure,
l Ischemia demonstrated by noninvasive testing in a patient with ejection fraction <40%, or ventricular arrhythmia
a prior myocardial infarction l Suspected or known mechanical complication post myocardial
l Post–cardiac transplant evaluation infarction
l Preoperation organ transplant in individuals 40 yr old l Myocardial infarction suspected to be from nonatherothrombotic
l Patients with equivocal or abnormal non–high risk findings on cause (i.e., spontaneous dissection, arteritis)
noninvasive testing with recurrent hospitalizations for chest pain l Cardiac trauma
l Suspected in-stent restenosis or early bypass graft failure on
the basis of symptoms or noninvasive testing
l Planned noncoronary cardiac surgery (valve surgery, hypertro-
phic cardiomyopathy, congenital heart disease)
l Before surgery for aortic aneurysm or dissection
l Unexplained systolic dysfunction
l Suspicion for ischemically mediated diastolic dysfunction
l Prospective immediate cardiac transplant donor at risk for cor-
onary disease
l Kawasaki disease with coronary artery aneurysms detected
noninvasively
Abbreviation: STEMI, ST segment elevation myocardial infarction.
Table 5.2 Relative Contraindications to Cardiac Catheterization consequences of the procedure” (5). The appropriateness criteria
apply to surgical and percutaneous coronary revascularization
. Severe contrast allergy
but do not address diagnostic cardiac catheterization.
. Acute renal failure or advanced chronic kidney disease without
the ability to perform hemodialysis In brief, indications for revascularization were developed
. Uncontrolled ventricular arrhythmia considering the following common variables: clinical presenta-
. Metabolic derangements (electrolytes, acid-base abnormalities) tion, angina severity, extent of ischemia on noninvasive testing,
. Drug toxicity (digoxin) the presence or absence of other prognostic factors, and the
. Malignant hypertension extent of anatomic disease. The technical panel scored each
. Active noncardiac problems (bacteremia, bleeding, stroke) indication on a scale from 1 to 9. Indications that were scored 7
. Acute decompensated heart failure in the absence of ischemia to 9 by the panel were termed appropriate, meaning coronary
. Coagulopathy or bleeding diathesis revascularization is generally acceptable, a reasonable
. Patient unable to cooperate with instructions
approach, and likely to improve the patient’s health outcomes
or survival. Indications with a score of 4 to 6 were termed
uncertain, meaning coronary revascularization may be accept-
Appropriateness Criteria able and a reasonable approach but with uncertainty implying
Marked variability in the use of coronary angiography and that more research and/or patient information is needed to
revascularization in and outside the United States have led to further classify the indication. Inappropriate indications for
concerns about the appropriate use of cardiovascular proce- coronary revascularization, score of 1 to 3, are generally not
dures (2–4). In addition to regional variation in cost, both acceptable and unlikely to improve the patient’s health out-
under- and overuse of invasive procedures have the potential comes or survival. In general, the technical panel rated the
to adversely impact patient outcomes. Appropriateness criteria majority of clinical scenarios in patients with acute coronary
for coronary revascularization were therefore developed in an syndromes and high-risk noninvasive findings as appropriate
effort to optimize the quality of cardiovascular care and for revascularization.
develop a tool that can be used to measure variability and Operators performing cardiac catheterization must be
utilization patterns in the use of coronary revascularization. In familiar with the appropriateness criteria. While the criteria
determining whether coronary revascularization was appropri- are not meant to replace clinical judgment, they do provide
ate, inappropriate or uncertain, the following definition was guidance for decision making and supplement practice guide-
used, “Coronary revascularization is appropriate when the expected lines developed by the ACC/AHA. Importantly, the criteria
benefits, in terms of survival or health outcomes (symptoms, func- will certainly be used by health care facilities and third-party
tional status, and/or quality of life) exceed the expected negative payers and ultimately patients.
[14/6/010/12:55:6] [38–46]
Table 5.3 Arterial Access Considerations

Femoral approach Brachial/radial approach
1. Requirement of a large sheath size (>7F) 1. Aortic pathology (dissection, aneurysms, large atheroma)
2. Simultaneous arterial and venous access planned 2. Peripheral arterial disease (aortic or iliac occlusions, iliac tortuosity)
3. Abnormal allen test or known subclavian/innominate artery 3. Morbid obesity
disease 4. Difficulty maintaining a supine position (back pain, lung disease)
5. Coagulopathy
PREPARATION use of uncommon conduits such as a gastroepiploic graft, or

Patient preparation is one of the most important aspects of placement of a saphenous vein graft off the descending aorta.
cardiac catheterization. Each patient has unique issues that When the operative report is unavailable and the procedure
must be addressed, for example, a history of contrast reaction, cannot be delayed, aortography and bilateral mammary injec-
coumadin therapy, or diabetes mellitus. In addition to clinical tions should be considered to assist in locating grafts.
factors, the physician must ensure that the patient is mentally
prepared, understands instructions, and knows what to expect
before, during, and after the procedure. A well informed
Access Considerations
Whether to use femoral, brachial, or radial access should be
patient will be less anxious and more cooperative. The circum-
considered prior to the procedure (Table 5.3). While many
stances of the procedure will dictate the pace of preparation,
operators and catheterization laboratories have a default strat-
but even in emergent situations such as ST segment elevation
egy of using the right femoral artery approach, patient and
myocardial infarction (STEMI) patient safety cannot be com-
lesion specific factors may obligate an alternative approach. A
promised by inadequate preprocedural assessment.
review of the patient’s comorbidities, need for left (vs. right and
left) heart catheterization, and anticipated maximum sheath
Preprocedural Assessment size can decrease the need for more than one access site. The
A detailed medical history, pertinent physical examination, and femoral approach should be considered for procedures requir-
review of allergies and ancillary studies should be performed ing a large sheath size, technical complexity, or simultaneous
within 30 days and updated the day of the procedure. In arterial and venous access. Radial access should be considered
patients with known or suspected coronary artery disease, in patients unable to lay flat for prolonged periods because of
assessment of compliance and ability to take dual antiplatelet pulmonary or musculoskeletal conditions, or patients with an
therapy should be made. Any upcoming noncardiac proce- increased risk of local vascular complications such as those
dures and surgeries should also be discussed with the patients. with peripheral artery disease, aortic dissection or aneurysm,
The electrocardiogram should be reviewed and available at the morbid obesity, or coagulopathy. In patients with a mammary
time of the procedure for comparison. In some individuals a graft, the ipsilateral side should be used if an arm approach is
chest X ray may be warranted. Basic laboratories, including a chosen to simplify selective mammary engagement. In a meta-
complete blood count, glucose, electrolytes, blood urea nitrogen analysis of randomized trials comparing radial with femoral
and creatinine should be obtained in all patients to assess the access that included 3224 patients, the risk of access site
status of chronic diseases or to identify clinically silent comor- complications was significantly less with the radial approach,
bidities such as kidney disease and anemia. Patients with a whereas the risk of major adverse cardiac events was similar.
personal or family history of a bleeding diathesis or antico- The radial approach, however, was associated with a higher
agulation should have coagulation studies. Abnormalities risk of procedural failure (6). Procedural success may depend
should be recorded and corrected if possible. The procedure, in part on operator experience with the radial technique.
if elective, should be delayed when problems are identified that The palmar arch should be assessed prior to radial artery
could influence treatment strategies, for example, unexplained cannulation because there is a 5% to 19% occurrence of radial
anemia. Abnormalities such as an elevated creatinine or throm- artery occlusion (7). In patients with an intact arch, extensive
bocytopenia may not alter plans for diagnostic cardiac catheter- collaterals between the radial and ulnar arteries make this
ization, but may require additional measures to reduce the risk complication clinically irrelevant. Hand ischemia can occur
of contrast nephropathy or bleeding, respectively. with radial artery occlusion in patients that have incomplete
For patients with prior cardiac catheterization, the palmar arches. The modified Allen test assesses the palmar
approach, equipment, and findings including a review of the arch and should be performed prior to radial artery catheter-
images should be performed whenever feasible. There is no use ization. The test is performed by compressing both the radial
trying to engage the left coronary artery with a Judkins left and ulnar arteries while the patients hand is held high with the
4 catheter when the previous operator was successful with a fist clenched. The hand is then lowered and opened and pres-
Judkins left 6 or to rediscover an anomalous coronary artery. sure over the ulnar artery released. If the superficial palmar
For patients with prior coronary bypass surgery (CABG), the arch is intact color should return to the hand within 6 seconds,
preoperative native coronary anatomy, operative report, and and 10 seconds is considered abnormal. Since the Allen test is
postoperative angiograms if performed, should be reviewed. subject to inaccuracy, many centers use pulse oximetry and
Every effort should be made to review the original operative plethysmography as a more direct assessment of blood flow
report rather than relying on a summary of the graft anatomy (8,9). Plethysmography is considered abnormal if during radial
which may be misleading. Attention should be paid to the artery compression there is a loss of pulse tracing with no
number and types of conduits and any special circumstances recovery during two minutes of observation. In 1010 consecu-
such as use of a free versus in situ left internal mammary graft, tive patients referred for cardiac catheterization, the modified
[14/6/010/12:55:6] [38–46]
Allen test was abnormal in 6.3% and plethysmography in 1.5%, approach may be preferable and vitamin K+ and fresh frozen
but the clinical implications of using a particular method are plasma can be transfused if bleeding complications occur. One
not clear (8). study suggested that manual compression is an acceptable
An abnormal Allen test is observed in 6% to 27% of method of hemostasis in therapeutically anticoagulant patients
patients undergoing cardiac catheterization (8,10). A small with an INR of 2 to 3 (15).
study of patients undergoing coronary angiography showed
that patients with an abnormal Allen test had decreased thumb
blood flow and increased capillary lactate after 30 minutes of Diabetes Mellitus
radial artery occlusion compared with patients with a normal Patients with diabetes treated with oral agents or insulin
Allen test (7). Given the potential risk of hand ischemia, the require an adjustment in medication to prevent periprocedural
radial artery approach should not be used in patients with an hypoglycemia while fasting. Patients taking subcutaneous insu-
abnormal modified Allen test unless the risks of alternative lin should be instructed to take half doses of long-acting insulin
approaches are prohibitory. preparations the morning of the procedure. Regular insulin
should be held and a blood glucose level checked by the
catheterization laboratory staff for symptoms of hypo- or
Conditions Requiring Special Preparations hyperglycemia. Oral agents should be held the morning of
Allergies the procedure as well. Patients with diabetes, particularly
A list of allergies should be recorded and patients should be those taking neutral protamine Hagedorn (NPH) insulin,
specifically questioned regarding allergies to lidocaine, pre- have an increased sensitivity to protamine, an agent used to
medications (i.e., valium), and contrast media. An alternative reverse systemic heparinization. Major reactions to protamine
local anesthetic, Marcaine (1 mg/mL), can be used in patients including vasomotor collapse occur 50-fold more frequently in
with an allergy to lidocaine. Sedatives can be withheld or NPH insulin-dependent diabetes patients, therefore this drug
alternatives substituted for patients with prior reactions to should be used with caution in this patient subset (16).
benzodiazepines or narcotics. The overall incidence of adverse Patients on metformin require special instructions and
reactions to intravascular contrast media is about 5%, but rates monitoring because of rare cases of metformin-associated lactic
are higher in patients with a history of allergy, asthma, or prior acidosis reported in patients with diabetes and chronic kidney
contrast media reaction (11,12). Patients reporting contrast disease (17). Metformin is excreted in the urine and 90% is
media reactions require premedication prior to cardiac cathe- eliminated within 24 hours, therefore it is contraindicated in
terization. Premedication should also be considered is in patients with elevated creatinine. Since contrast studies can
patients with severe asthma or prior anaphylaxis. Pretreatment induce acute renal failure, there is a small risk of lactic acidosis
with corticosteroids, when given at least 12 hours before the in patients undergoing contrast administration. Nearly all the
procedure, significantly decreased contrast reactions with the cases reported, however, were in individuals with preexistent
exception of hives which had only a trend toward reduction. A renal impairment who were continued on the drug after con-
single-dose regimen administered two hours prior to contrast trast exposure (18). Patients with a creatinine <1.5 mg/dL can
exposure, however, did not reduce the incidence of reactions undergo cardiac catheterization without discontinuing metfor-
(13). On the basis of the limited data available, nonemergent min before the study. Metformin should be held post procedure
procedures should be delayed until adequate premedication and resumed in 48 hours in patients without signs of renal
can be administered. An acceptable regimen is oral prednisone failure. In patients at high risk of renal failure due to concom-
40 mg every 6 hours starting at least 12 and up to 24 hours itant mediation such as cyclosporine, volume depletion, or low
before procedure (see chap. 4 for different dosing schedule). output, the creatinine should be measured prior to resuming
For emergent circumstances intravenous solumedrol can be metformin. In patients with an elevated creatinine, metformin
given despite the lack of known efficacy. In addition to steroids, should be discontinued 48 hours prior to an elective procedure.
50 mg of oral or intravenous diphenhydramine should be In emergent situations, metformin should be discontinued at
given before the procedure. Compared with high-osmolar the time of the procedure, hydration administered, and renal
ionic contrast, there is a much lower incidence of adverse function monitored closely (19).
reaction to the low-osmolar nonionic contrast agents primarily
used today (14).
Renal Disease
Anticoagulants Chronic kidney disease, diabetes, low cardiac output, hypovo-
Patients on oral anticoagulant therapy should be assessed for lemia, and contrast volume are risk factors for contrast nephr-
the ability to temporarily hold warfarin to decrease the bleed- opathy. Patients with baseline renal insufficiency that have
ing risk associated with cardiac catheterization. When the risk deterioration in renal function after percutaneous coronary
of a subtherapeutic international normalized ratio (INR) is intervention (PCI) have poor outcomes, particularly those
acceptable, warfarin can be held for several days to allow the requiring hemodialysis (20). Patients at risk for contrast nephr-
INR optimally to drift down to 1.5. For high-risk clinical opathy should be adequately hydrated with intravenous saline
situations such as recent pulmonary embolism, atrial fibrilla- for 12 hours before procedure, if possible (21,22). Alternative
tion with prior embolic events, or mechanical mitral valve pretreatment regimens include oral N-acetylcysteine 600 mg
prosthesis, patients require periprocedural heparin to bridge every 12 hours starting 24 to 48 hours before the procedure or
the discontinuation of warfarin. Low-molecular-weight heparin sodium bicarbonate infusion 3 mL/kg/hr for 1 hour prior to
can be administered on an outpatient basis or the patient can be contrast administration followed by an infusion of 1 mL/kg/hr
admitted for intravenous unfractionated heparin. In some for 6 hours after the procedure (23,24). Additional measures
circumstances, catheterization must be performed while a including holding diuretics and other nephrotoxins such as
patient is fully anticoagulated. For these cases, a radial nonsteroidal anti-inflammatory agents should be considered.
[14/6/010/12:55:6] [38–46]
Table 5.4 Major Complications of Diagnostic Left Heart Catheter- Myocardial Infarction
ization Myocardial ischemia may occur during diagnostic angiography
as a result of catheter engagement or contrast injection and
Complication Incidence (%) References
a
more commonly during PCI with balloon inflations, but is
Mortality .08–.14 27,28,30,31 usually transient and self-limited. Development of anginal
Myocardial infarction .05–.17 27,30,35 symptoms, electrocardiographic, or hemodynamic changes
Stroke .18–.4 38–42
may all indicate ischemia and should prompt an evaluation
Vascular-related requiring .5 35
treatment of the cause. Attention to catheter flushing and monitoring for
Ventricular tachycardia 0.4 27 pressure damping reduces the risk of ischemia due to technical-
a related factors such as embolization or vessel occlusion. Even
Increased risk in patients with left main disease greater than 50%,
with careful technique, removal of the catheter from the coro-
ejection fraction less than 30%; NYHA class III or IV, age over 60 years,
three-vessel disease, valvular heart disease, renal disease. nary ostium or administration of nitroglycerin may be required
to reverse ischemia due to obstructive coronary artery disease
or vasospasm.
RISKS During diagnostic catheterization, progression of ische-
The risk of complications from cardiac catheterization has mia to MI is uncommon. In the first, second, and third SCAI
decreased over time and is <1% in the majority of cases registries, the risk of MI was rare and decreased over time, from
(Table 5.4). Even with experienced operators and modern 0.07% and 0.06% to 0.05% (28,31,36). Patient-related factors
equipment, complications will occur. Many factors including such as the extent of coronary artery disease (0.06% for single-
patient demographics, comorbidities, cardiovascular anatomy, vessel disease, 0.08% for triple-vessel disease) and location
procedural type and circumstances, and operator and hospital (0.17% for left main disease) slightly increase the risk of MI
volume, can influence the risk to an individual patient and (28). Postprocedural MI is rare following an uncomplicated
these should be weighed in each case (25,26). The risks are elective diagnostic catheterization and the majority of patients
lowest in stable patients undergoing elective procedures and can be discharged after several hours of observation (37).
diagnostic angiography in this setting has <0.1% risk of a major
adverse event such as death, MI, or stroke (27–29). The risk can
increase up to eightfold in patients with multivessel disease,
congestive heart failure, and renal insufficiency (30). To justify
Cerebrovascular Events
Stroke is an uncommon but potentially devastating complica-
performance of the procedure the expected benefits, in terms of
tion of cardiac catheterization and PCI. In several large series
diagnostic or therapeutic outcomes, should be greater than the
involving over 43,000 patients, the rate of clinically diagnosed
potential risks. Physicians performing invasive procedures
procedural related stroke ranged from 0.18% to 0.4% (38–42).
have to be knowledgeable about potential complications and
Several patient and procedural factors are associated with an
capable of administering treatment. This way every effort can
increased risk of stroke, including diabetes, prior stroke, longer
be made to minimize the incidence and severity of complica-
procedure times, intra-aortic balloon pump placement, and
tions, many of which are immediately life-threatening. The
treatment with fibrinolytic therapy (40,43). Patients that suffer
most common complications of cardiac catheterization are
from a periprocedural stroke have a poor prognosis with per-
reviewed below.
sistent neurologic deficits and a high in-hospital mortality rate
of up to 32% (38).
Death Cerebral microembolism is the primary mechanism of
Despite an aging population and performance of coronary periprocedural ischemic stroke occurring with left heart cath-
angiography in higher risk individuals, the procedural mortal- eterization. Manipulation of guidewires and catheters results
ity reported for 222,553 patients by the Society for Cardiac in disruption of atheromatous plaques from the walls of the
Angiography and Interventions (SCAI) registry in 1989 was aorta. In more than 50% of PCIs, guiding catheter placement is
0.098%, compared with 0.14% reported in 1982 (28,29). Proce- associated with scraping debris from the aorta as indicated by
dure-related mortality in a subsequent multicenter registry of retrieval of atheromatous material (44). Transcranial Doppler
58,332 patients in 1990 was 0.08% (31). A later single-center and serial magnetic resonance imaging studies of patients
study of 11,821 patients undergoing 7953 diagnostic and 3868 undergoing cardiac catheterization and PCI support emboliza-
therapeutic procedures was congruent. The total mortality rate tion as the main cause of stroke with asymptomatic micro-
was 0.2%, with a fivefold higher rate for interventions, com- emboli detected in about 15% of cases (45–47). Stroke can also
pared with diagnostic procedures (32). The cause of in-hospital result from embolization of air or thrombus, or intracerebral
death after PCI has been evaluated and procedural complica- hemorrhage. Stroke risk can be minimized, therefore, by
tions accounted for about half of deaths and the remaining meticulous procedural technique such as withdrawal of
cases were attributed to preexisting cardiac disease, predom- blood from and flushing of catheters with heparinized saline
inantly low output failure (33). and performing catheter exchanges over wires in the descend-
Several subgroups at higher risk of procedural mortality ing aorta.
have been identified. In the initial registry, patients with left The optimal management of periprocedural stroke is
main disease >50%, ejection fraction <30%; NYHA class III or unknown, but favorable results have been shown in small
IV, age >60 years, and three-vessel disease were at increased series of patients treated with neurovascular intervention and
risk of mortality from diagnostic cardiac catheterization (34). intraarterial fibrinolytics (48–50). The clinical situation should
Renal insufficiency, postprocedural renal function deteriora- prompt an emergent response and multidisciplinary manage-
tion, and valvular heart disease are additional predictors of ment with cardiology, neurology and a neurointerventional
mortality (20,35). specialist (51).
[14/6/010/12:55:6] [38–46]
Vascular Complications biopsy, or elective pericardiocentesis. Patients may present

Several complications can occur at the arterial site of catheter with vagal reactions, hypotension, chest pain, or arrhythmia.
insertion that cause significant patient discomfort, increased Echocardiographic guidance can be used if immediately avail-
length of stay, and increased health care costs. Vascular access able and anticoagulation should be reversed if possible. In a
site complications are not uncommon, but only about 0.5% of large single-center series, emergent pericardiocentesis relieved
cases require specific intervention. The majority of these com- acute tamponade in 99% of patients and was the only therapy
plications, including arterial thrombosis, dissection, uncon- required in 82% of patients (59).
trolled or retroperitoneal bleeding, and hematoma formation,
commonly occur during or within hours following catheter- Atheroembolism
ization. Development of a pseudoaneurysm or arteriovenous Atheromatous debris can be dislodged from the arterial wall
fistula, however, may not manifest for several days. In general, during catheter manipulation and exchanges and may result in
arterial thrombosis is more common with radial access, systemic embolization. When systematically evaluated, approx-
whereas other vascular complications are more frequent with imately half of patients undergoing PCI had detectable athe-
femoral or brachial approaches (6,36). Treatment of vascular romatous debris in blood removed after placement of the
complications is often conservative for hematomas but surgical guiding catheter (44). None of the patients, however, had
intervention or interventional techniques are required for arte- clinical events, suggesting that operator technique is critical
rial thrombosis, arteriovenous fistulas, large or expanding to the prevention of embolic events. The incidence of clinically
pseudoaneurysms, and retroperitoneal bleeds not responsive evident atheroembolic events is reported to be 0.6% to 1.9%
to supportive care (52). (60). In a large prospective study, cholesterol embolization
Hemostasis at the access site can be accomplished either syndrome, as defined by peripheral cutaneous involvement
by manual or device compression or with a closure device, of or renal dysfunction, occurred in 1.4%; 48% of patients had
which there are several types (53–55). The approach chosen for cutaneous involvement as defined by the presence of livedo
hemostasis depends on catheterization laboratory and patient- reticularis, blue toe syndrome, or digital gangrene; and 64% of
related factors such as availability of trained staff for sheath patients had renal insufficiency. Eosinophil counts were sig-
removal, anticoagulation status, sheath size and location, pres- nificantly higher in patients suffering from atheroembolism
ence of peripheral vascular disease, and patient comfort. In both before and after catheterization, but the only independent
patients undergoing diagnostic procedures, collagen and predictor of cholesterol emboli syndrome was baseline C-
suture type arterial puncture closure devices shorten the reactive protein. Unadjusted mortality was 32-fold greater in
time to ambulation and have comparable safety to manual patients with cholesterol emboli syndrome (16% vs. 0.5%).
compression (54–56). In two large meta-analyses involving Unfortunately, no specific therapy exists and care is supportive,
over 37,000 patients undergoing diagnostic and interventional with hemodialysis and wound management as required (61).
procedures with numerous types of devices in 30 studies,
closure devices had an increased risk of local complications Infection
(56,57). The risk of complications, however, differed among the As cardiac catheterization is performed using sterile technique,
various closure devices (56). Overall, the data suggest that the incidence of bacteremia or infection at the access site is rare
closure devices are an acceptable alternative to manual com- and routine use of prophylactic antibiotics is not recom-
pression in appropriate patients. mended. SCAI infection control guidelines discuss techniques
for minimizing risk to the patient and catheterization labora-
Arrhythmias tory personnel (62). Protection of both the patient and person-
Although predominantly benign and self-limited, a gamut of nel includes the proper use of gowns, caps, gloves, masks, and
arrhythmias can occur during cardiac catheterization, necessi- sterile technique including hand washing. Additional protec-
tating continuous electrocardiographic monitoring and a staff tion to laboratory staff comes from eye wear and proper
trained in arrhythmia treatment. The majority of arrhythmias handling and disposal of contaminated equipment. All person-
are induced by intracoronary contrast injection or right heart nel should receive vaccination for hepatitis B (63). Immediate
catheter manipulation. Ventricular tachycardia and fibrillation cleansing of needlestick injuries and prompt medical attention
are rare, but occurred in 0.4% of cases in the second SCAI is required. The use of prophylactic antibiotics is recommended
registry (28). Bradycardia and conduction disturbances are also by the manufacturers of certain closure devices.
uncommon, with an incidence of approximately 1% in diag-
nostic catheterization and PCI. Patients are generally respon-
sive to forceful coughing and atropine. Initiation of temporary
Radiation Exposure
Although the risk of radiation to patients and staff is an
pacing is rarely required (0.06% diagnostic cases, 0.4% PCI)
accepted consequence of cardiac catheterization, every effort
(58). Vasovagal reactions occur in up to 3% of patients as a
should be made to manage patient radiation dose (64). All
result of anxiety or pain. Patients generally respond to removal
personnel working with fluoroscopy should be trained and
of painful stimuli, fluid administration, and atropine. These
competency monitored (65). Radiation exposure is determined
reactions, however, can be life-threatening in patients with
by multiple factors (Table 5.5) (66,67). Cardiac catheterization
severe valvular or ischemic disease and should be rapidly
laboratory systems should be equipped with the capability of
treated to prevent hemodynamic decompensation.
monitoring cumulative radiation dose. The energy delivered to
air by the X-ray beam, or air kerma, is measured at a fixed
Perforation reference point during a procedure to calculate the reference
The risk of perforation of the heart or blood vessels is exceed- point dose. The reference point dose is the cumulative dose
ingly rare in diagnostic cardiac catheterization, but may occur derived from fluoro, cine, and dosimetric effects of patient size
with transseptal procedures, pacemaker placement, myocardial and is typically displayed as milligray (64). Some systems use a
[14/6/010/12:55:6] [38–46]
Table 5.5 Factors That Contribute to Patient Radiation Exposure CONCLUSIONS

To provide the highest level of care with the lowest risk,
. X-ray system
invasive and interventional cardiologists must ensure they
. Body mass index/weight distribution of the patient
have up-to-date, in-depth knowledge of procedural indications
. Duration of fluoroscopy
. Cine usage and risks and meticulous technical skills. Appropriate patient
. Table position (distance between patient and X-ray tube) selection and preparation requires individualized care and
. Technique selection [i.e., magnification, fluoroscopy and cine attention to high-risk features. Procedural risk and patient
modes (frames per second), beam angulation] stress can be minimized by the process of informed consent,
patient preparation, and an experienced, trained catheterization
laboratory staff.
kerma area product meter, or dose area product meter, to

monitor dose. REFERENCES
1. Scanlon PJ, Faxon DP, Audet AM, et al. ACC/AHA guidelines for
The most common type of radiation injury to patients is
coronary angiography: executive summary and recommenda-
skin burns. All patients receiving a skin dose of 5 to 10 Gy will tions. A report of the American College of Cardiology/American
have erythema. Obese patients and those with long fluoroscopy Heart Association Task Force on Practice Guidelines (Committee
times are at highest risk. The rash is located at the radiation on Coronary Angiography) developed in collaboration with the
entry portal which is usually on the back. When institutional Society for Cardiac Angiography and Interventions. Circulation
threshold for a significant dose are reached, a follow-up process 1999; 99:2345–2357.
should be initiated. Patients at risk should be instructed to 2. Guadagnoli E, Landrum MB, Normand SL, et al. Impact of
survey their skin at 48 to 72 hours for early signs or transient underuse, overuse, and discretionary use on geographic variation
erythema. Since reactions may be delayed, surveillance for skin in the use of coronary angiography after acute myocardial infarc-
changes should continue for four to six weeks following tion. Med Care 2001; 39:446–458.
3. Hartford K, Ross LL, Walld R. Regional variation in angiography,
prolonged radiation exposure. The Joint Commission sentinel
coronary artery bypass surgery, and percutaneous transluminal
event threshold is 15 Gy and doses above this will result in coronary angioplasty in Manitoba, 1987 to 1992: the funnel effect.
deep tissue injury. The effect of radiation exposure on cancer Med Care 1998; 36:1022–1032.
risk is difficult to determine, but it should be recognized that 4. Hannan EL, Kumar D. Geographic variation in the utilization and
diagnostic X rays are the largest man-made source of radiation choice of procedures for treating coronary artery disease in New
exposure and do contribute to cancer risk (68). York State. Ischaemic Heart Disease Patient Outcomes Research
Team (PORT). J Health Serv Res Policy 1997; 2:137–143.
5. Patel MR, Dehmer GJ, Hirshfeld JW, et al. ACCF/SCAI/STS/
INFORMED CONSENT AATS/AHA/ASNC 2009 Appropriateness Criteria for Coronary
The process of obtaining informed consent is a critical aspect of Revascularization: a report by the American College of Cardiology
cardiac catheterization and intervention. Responsibility ulti- Foundation Appropriateness Criteria Task Force, Society for
mately lies with the operating physician. While members of Cardiovascular Angiography and Interventions, Society of
Thoracic Surgeons, American Association for Thoracic Surgery,
the catheterization laboratory team may take part in describing
American Heart Association, and the American Society of Nuclear
the flow and aspects of the procedure, the physician should Cardiology Endorsed by the American Society of Echocardiogra-
confirm that the risks, benefits, and alternative treatments to phy, the Heart Failure Society of America, and the Society of
the planned procedure have been explained to the patient’s Cardiovascular Computed Tomography. J Am Coll Cardiol 2009;
satisfaction. Sufficient detail should be provided to adequately 53:530–553.
inform the patient without provoking undue fear. The patient 6. Agostoni P, Biondi-Zoccai GGL, de Benedictis ML, et al. Radial
or patient’s legal representative should provide written consent versus femoral approach for percutaneous coronary diagnostic
and documentation should be placed in the medical record. In and interventional procedures; Systematic overview and meta-
rare circumstances when a patient is unable to provide consent analysis of randomized trials. J Am Coll Cardiol 2004; 44:349–356.
and delay would compromise survival (such as with a STEMI), 7. Greenwood MJ, Della-Siega AJ, Fretz EB, et al. Vascular commu-
nications of the hand in patients being considered for transradial
medical justification for proceeding without consent should be
coronary angiography: is the Allen’s test accurate [see comment]?
documented in the medical record. Hospital personnel such as J Am Coll Cardiol 2005; 46:2013–2017.
social workers can be of assistance in locating family members 8. Barbeau GR, Arsenault F, Dugas L, et al. Evaluation of the
while care is being provided to the patient. The rationale for ulnopalmar arterial arches with pulse oximetry and plethysmog-
proceeding with emergent procedures should be explained to raphy: comparison with the Allen’s test in 1010 patients. Am
the patient and family as soon as feasible. Heart J 2004; 147:489–493.
Another special circumstance involving consent pertains 9. Hovagim AR, Katz RI, Poppers PJ. Pulse oximetry for evaluation
to medical research. Research protocols that involve patients in of radial and ulnar arterial blood flow. J Cardiothorac Anesth
the cardiac catheterization laboratory such as studies involving 1989; 3:27–30.
drugs or devices should be thoroughly explained to the patient 10. Benit E, Vranckx P, Jaspers L, et al. Frequency of a positive
modified Allen’s test in 1,000 consecutive patients undergoing
prior to the initiation of the procedure. Patients must be given
cardiac catheterization. Cathet Cardiovasc Diagn 1996; 38:352–354.
time to fully examine the research protocol consent form and 11. Shehadi WH, Toniolo G. Adverse reactions to contrast media: a
must understand the protocol including treatments, risks, and report from the Committee on Safety of Contrast Media of the
follow-up. Patients should not be pressured to consent to International Society of Radiology. Radiology 1980; 137:299–302.
research protocols. Physicians should disclose their relation- 12. Lang DM, Alpern MB, Visintainer PF, et al. Increased risk for
ship to the study. The ethical principles and guidelines for the anaphylactoid reaction from contrast media in patients on beta-
protection of human subjects of research should be followed adrenergic blockers or with asthma [see comment]. Ann Intern
(69,70). Med 1991; 115:270–276.
[14/6/010/12:55:6] [38–46]
13. Lasser EC, Berry CC, Talner LB, et al. Pretreatment with cortico- 32. Chandrasekar B, Doucet S, Bilodeau L, et al. Complications of
steroids to alleviate reactions to intravenous contrast material. N cardiac catheterization in the current era: a single-center experi-
Engl J Med 1987; 317:845–849. ence [see comment]. Cathet Cardiovasc Interv 2001; 52:289–295.
14. Katayama H, Yamaguchi K, Kozuka T, et al. Adverse reactions to 33. Malenka DJ, O’Rourke D, Miller MA, et al. Cause of in-hospital
ionic and nonionic contrast media. A report from the Japanese death in 12,232 consecutive patients undergoing percutaneous
Committee on the Safety of Contrast Media [see comment]. Radi- transluminal coronary angioplasty. The Northern New England
ology 1990; 175:621–628. Cardiovascular Disease Study Group [see comment]. Am Heart J
15. El-Jack SS, Ruygrok PN, Webster MWI, et al. Effectiveness of 1999; 137:632–638.
manual pressure hemostasis following transfemoral coronary 34. Kennedy JW, Baxley WA, Bunnel IL, et al. Mortality related to
angiography in patients on therapeutic warfarin anticoagulation. cardiac catheterization and angiography. Cathet Cardiovasc
Am J Cardiol 2006; 97:485–488. Diagn 1982; 8:323–340.
16. Stewart WJ, McSweeney SM, Kellett MA, et al. Increased risk of 35. Folland ED, Oprian C, Giacomini J, et al. Complications of cardiac
severe protamine reactions in NPH insulin-dependent diabetics catheterization and angiography in patients with valvular heart
undergoing cardiac catheterization. Circulation 1984; 70:788–792. disease. VA Cooperative Study on Valvular Heart Disease [see
17. Gan SC, Barr J, Arieff AI, et al. Biguanide-associated lactic acido- comment]. Cathet Cardiovasc Diagn 1989; 17:15–21.
sis. Case report and review of the literature. Arch Intern Med 36. Kennedy JW. Complications associated with cardiac catheteriza-
1992; 152:2333–2336. tion and angiography. Cathet Cardiovasc Diagn 1982; 8:5–11.
18. Bailey CJ, Turner RC. Metformin [see comment]. N Engl J Med 37. Mahrer PR, Young C, Magnusson PT. Efficacy and safety of out-
1996; 334:574–579. patient cardiac catheterization. Cathet Cardiovasc Diagn 1987;
19. Heupler FA Jr. Guidelines for performing angiography in patients 13:304–308.
taking metformin. Members of the Laboratory Performance Stan- 38. Fuchs S, Stabile E, Kinnaird TD, et al. Stroke complicating percu-
dards Committee of the Society for Cardiac Angiography and taneous coronary interventions: incidence, predictors, and prog-
Interventions. Cathet Cardiovasc Diagn 1998; 43:121–123. nostic implications. Circulation 2002; 106:86–91.
20. Gruberg L, Mintz GS, Mehran R, et al. The prognostic implications 39. Lazar JM, Uretsky BF, Denys BG, et al. Predisposing risk factors
of further renal function deterioration within 48 h of interven- and natural history of acute neurologic complications of left-sided
tional coronary procedures in patients with pre-existent chronic cardiac catheterization. Am J Cardiol 1995; 75:1056–1060.
renal insufficiency. J Am Coll Cardiol 2000; 36:1542–1548. 40. Segal AZ, Abernethy WB, Palacios IF, et al. Stroke as a complica-
21. Trivedi HS, Moore H, Nasr S, et al. A randomized prospective trial tion of cardiac catheterization: risk factors and clinical features.
to assess the role of saline hydration on the development of contrast Neurology 2001; 56:975–977.
nephrotoxicity [see comment]. Nephron 2003; 93:C29–C34. 41. Wong SC, Minutello R, Hong MK. Neurological complications
22. Mueller C, Buerkle G, Buettner HJ, et al. Prevention of contrast following percutaneous coronary interventions (a report from the
media-associated nephropathy: randomized comparison of 2000–2001 New York State Angioplasty Registry). Am J Cardiol
2 hydration regimens in 1620 patients undergoing coronary 2005; 96:1248–1250.
angioplasty [see comment]. Arch Intern Med 2002; 162:329–336. 42. Dukkipati S, O’Neill WW, Harjai KJ, et al. Characteristics of
23. Merten GJ, Burgess WP, Gray LV, et al. Prevention of contrast- cerebrovascular accidents after percutaneous coronary interven-
induced nephropathy with sodium bicarbonate: a randomized tions [see comment]. J Am Coll Cardiol 2004; 43:1161–1167.
controlled trial [see comment]. JAMA 2004; 291:2328–2334. 43. Sutton AGC, Campbell PG, Graham R, et al. A randomized trial of
24. Pannu N, Manns B, Lee H, et al. Systematic review of the impact rescue angioplasty versus a conservative approach for failed
of N-acetylcysteine on contrast nephropathy [see comment]. fibrinolysis in ST-segment elevation myocardial infarction: the
Kidney Int 2004; 65:1366–1374. Middlesbrough Early Revascularization to Limit INfarction
25. Baim D, Grossman W. Complications of cardiac catheterization. (MERLIN) trial [see comment]. J Am Coll Cardiol 2004; 44:
In: Baim D, Grossman W, eds. Cardiac Catheterization, Angiog- 287–296.
raphy and Intervention. 6th ed. Baltimore: Williams & Wilkins, 44. Keeley EC, Grines CL. Scraping of aortic debris by coronary
2000:35–65. guiding catheters: a prospective evaluation of 1,000 cases [see
26. Hannan EL, Wu C, Walford G, et al. Volume-outcome relation- comment]. J Am Coll Cardiol 1998; 32:1861–1865.
ships for percutaneous coronary interventions in the stent era [see 45. Hamon M, Gomes S, Oppenheim C, et al. Cerebral microembolism
comment]. Circulation 2005; 112:1171–1179. during cardiac catheterization and risk of acute brain injury: a
27. Davis K, Kennedy JW, Kemp HG Jr., et al. Complications of prospective diffusion-weighted magnetic resonance imaging
coronary arteriography from the Collaborative Study of Coronary study. Stroke 2006; 37:2035–2038.
Artery Surgery (CASS). Circulation 1979; 59:1105–1112. 46. Leclercq F, Kassnasrallah S, Cesari JB, et al. Transcranial Doppler
28. Johnson LW, Lozner EC, Johnson S, et al. Coronary arteriography detection of cerebral microemboli during left heart catheterization.
1984–1987: a report of the Registry of the Society for Cardiac Cerebrovascular Diseases 2001; 12:59–65.
Angiography and Interventions. I. Results and complications. 47. Busing KA, Schulte-Sasse C, Fluchter S, et al. Cerebral infarction:
Cathet Cardiovasc Diagn 1989; 17:5–10. incidence and risk factors after diagnostic and interventional
29. Lozner EC, Johnson LW, Johnson S, et al. Coronary arteriography cardiac catheterization–prospective evaluation at diffusion-
1984–1987: a report of the Registry of the Society for Cardiac weighted MR imaging. Radiology 2005; 235:177–183.
Angiography and Interventions. II. An analysis of 218 deaths 48. Al-Mubarak N, Vitek JJ, Mousa I, et al. Immediate catheter-based
related to coronary arteriography. Cathet Cardiovasc Diagn 1989; neurovascular rescue for acute stroke complicating coronary
17:11–14. procedures. Am J Cardiol 2002; 90:173–176.
30. Laskey W, Boyle J, Johnson LW. Multivariable model for predic- 49. De Marco F, Antonio Fernandez-Diaz J, Lefevre T, et al. Manage-
tion of risk of significant complication during diagnostic cardiac ment of cerebrovascular accidents during cardiac catheterization:
catheterization. The Registry Committee of the Society for Cardiac immediate cerebral angiography versus early neuroimaging strat-
Angiography & Interventions. Cathet Cardiovasc Diagn 1993; egy. Cathet Cardiovasc Interv 2007; 70:560–568.
30:185–190. 50. Khatri P, Taylor RA, Palumbo V, et al. The safety and efficacy of
31. Noto TJ Jr., Johnson LW, Krone R, et al. Cardiac catheterization thrombolysis for strokes after cardiac catheterization [see com-
1990: a report of the Registry of the Society for Cardiac Angiog- ment]. J Am Coll Cardiol 2008; 51:906–911.
raphy and Interventions (SCA&I). Cathet Cardiovasc Diagn 1991; 51. Hamon M, Baron J-C, Viader F, et al. Periprocedural stroke and
24:75–83. cardiac catheterization. Circulation 2008; 118:678–683.
[14/6/010/12:55:6] [38–46]
52. Webber GW, Jang J, Gustavson S, et al. Contemporary manage- prospective study [see comment]. J Am Coll Cardiol 2003; 42:
ment of postcatheterization pseudoaneurysms. Circulation 2007; 211–216.
115:2666–2674. 62. Chambers CE, Eisenhauer MD, McNicol LB, et al. Infection control
53. Pracyk JB, Wall TC, Longabaugh JP, et al. A randomized trial of guidelines for the cardiac catheterization laboratory: society
vascular hemostasis techniques to reduce femoral vascular compli- guidelines revisited. Cathet Cardiovasc Interv 2006; 67:78–86.
cations after coronary intervention. Am J Cardiol 1998; 81:970–976. 63. Heupler FA Jr., Heisler M, Keys TF, et al. Infection prevention
54. Ward SR, Casale P, Raymond R, et al. Efficacy and safety of a guidelines for cardiac catheterization laboratories. Society for
hemostatic puncture closure device with early ambulation after Cardiac Angiography and Interventions Laboratory Performance
coronary angiography. Angio-Seal Investigators. Am J Cardiol Standards Committee. Cathet Cardiovasc Diagn 1992; 25:260–263.
1998; 81:569–572. 64. Balter S, Moses J. Managing patient dose in interventional cardi-
55. Baim DS, Knopf WD, Hinohara T, et al. Suture-mediated closure ology [see comment]. Cathet Cardiovasc Interv 2007; 70:244–249.
of the femoral access site after cardiac catheterization: results of 65. Hirshfeld JW Jr., Balter S, Brinker JA, et al. ACCF/AHA/HRS/
the suture to ambulate aNd discharge (STAND I and STAND II) SCAI clinical competence statement on physician knowledge to
trials. Am J Cardiol 2000; 85:864–869. optimize patient safety and image quality in fluoroscopically
56. Nikolsky E, Mehran R, Halkin A, et al. Vascular complications guided invasive cardiovascular procedures. A report of the
associated with arteriotomy closure devices in patients undergo- American College of Cardiology Foundation/American Heart
ing percutaneous coronary procedures: a meta-analysis [see com- Association/American College of Physicians Task Force on Clinical
ment]. J Am Coll Cardiol 2004; 44:1200–1209. Competence and Training. J Am Coll Cardiol 2004; 44:2259–2282.
57. Koreny M, Riedmuller E, Nikfardjam M, et al. Arterial puncture 66. Laskey WK, Wondrow M, Holmes DR, Jr. Variability in fluoro-
closing devices compared with standard manual compression scopic X-ray exposure in contemporary cardiac catheterization
after cardiac catheterization: systematic review and meta-analysis. laboratories. J Am Coll Cardiol 2006; 48:1361–1364.
JAMA 2004; 291:350–357. 67. Cusma JT, Bell MR, Wondrow MA, et al. Real-time measurement
58. Harvey JR, Wyman RM, McKay RG, et al. Use of balloon flotation of radiation exposure to patients during diagnostic coronary
pacing catheters for prophylactic temporary pacing during diag- angiography and percutaneous interventional procedures. J Am
nostic and therapeutic catheterization procedures. Am J Cardiol Coll Cardiol 1999; 33:427–435.
1988; 62:941–944. 68. Berrington de Gonzalez A, Darby S. Risk of cancer from diagnostic
59. Tsang TS, Freeman WK, Barnes ME, et al. Rescue echocardio- X-rays: estimates for the UK and 14 other countries [see comment].
graphically guided pericardiocentesis for cardiac perforation com- Lancet 2004; 363:345–351.
plicating catheter-based procedures. The Mayo Clinic experience. 69. Code of federal regulations regarding protection of human sub-
J Am Coll Cardiol 1998; 32:1345–1350. jects (45 CFR 46). Available at: http://ohsr.od.nih.gov/guidelines/
60. Bashore TM, Gehrig T. Cholesterol emboli after invasive cardiac 45cfr46.html. Accessed August 5, 2009.
procedures [comment]. J Am Coll Cardiol 2003; 42:217–218. 70. The Belmont report ethical principles and guidelines for the
61. Fukumoto Y, Tsutsui H, Tsuchihashi M, et al. Cholesterol Embo- protection of human subjects of research. Available at: http://
lism Study I. The incidence and risk factors of cholesterol embo- ohsr.od.nih.gov/guidelines/belmont.html. Accessed August 6,
lization syndrome, a complication of cardiac catheterization: a 2009.
[16/6/010/10:12:11] [47–52]
Conscious sedation (local anesthetics,

sedatives, and reversing agents)
Steven P. Dunn
INTRODUCTION vital signs and auscultation of the heart and lungs (1). Although
Conscious sedation, also known as moderate sedation, is the many catheterization laboratories achieve adequate preproce-
practice of choice recommended by the American College of dural sedation with the use of oral medications, intravenous
Cardiology (ACC) and the Society for Cardiac Angiography access for medication administration is a necessity.
and Interventions (SCAI) to relieve pain and anxiety during While there are many metabolic determinants to both the
cardiac catheterization (1). The American Society of Anesthesi- choice and the amount of sedation administered to a patient to
ologists (ASA) has defined moderate sedation as a drug- achieve conscious sedation, a systematic approach to both
induced depression of consciousness that continues to allow assessing the patient and selecting the therapeutic modality
the patient to purposefully respond to commands and maintain will ensure optimal outcomes.
airway, breathing, and circulation without mechanical or phar-
macologic support (Table 6.1) (2). This level of sedation is
optimal for cardiac catheterization since patient symptoms
Monitoring of Sedation
The ASA focuses on four key areas of monitoring for moderate
can often be the first warning of a procedural complication.
sedation: level of consciousness, pulmonary ventilation, oxy-
In addition, by allowing patients to maintain physiologic respi-
genation, and hemodynamics (1). Although the methods uti-
ratory and cardiac function, recovery time is significantly
lized to achieve moderate sedation may differ, the following
lessened. While there are general recommendations regarding
should serve as a universal method to assess whether appro-
the most appropriate methods to achieve this level of sedation,
priate sedation is achieved. Monitoring will require a combi-
significant variability in practice exists across different cardiac
nation of direct observation and assessment, along with various
catheterization laboratories.
medical equipment, as appropriate. Of note, as respiratory
depression is the principle concern with the use of moderate
FUNDAMENTALS sedative modalities, monitoring of both ventilation and oxy-
Preprocedure Assessment genation is recommended.
A preprocedure assessment is an important part of ensuring The level of sedation can also be objectively assessed with
safe and effective sedation therapy is applied to each patient the use of the Aldrete score (Table 6.4) (4). This score assigns
situation, as many metabolic determinants can put the patient point values for activity, respiration, circulation, consciousness,
at risk for adverse reaction to sedation (Table 6.2) (1). Ideally and color.
this would include an assessment of major organ system func-
tion, history of drug and alcohol abuse (including tobacco), Level of Consciousness
date and time of last oral intake, and allergic history relative to For moderate sedation, a level of consciousness where the
local anesthesia and sedative drugs. patient purposefully responds to commands (physically or
As sedative agents can affect airway reflexes, patients verbally) is appropriate. While there is little literature to sug-
undergoing cardiac catheterization are at risk for aspiration of gest that monitoring level of consciousness improves clinical
gastric contents. Therefore, patients undergoing this procedure outcomes associated with procedural sedation, it is strongly felt
should ideally fast for an appropriate amount of time to allow that early intervention when the patient’s level of sedation is
for sufficient gastric emptying, generally four hours after a supertherapeutic (e.g., no response, or response only to painful
meal. Clear liquids may be permitted up to two hours prior to stimuli) will likely prevent adverse outcomes associated with
the procedure (2). Medications, particularly antihypertensive sedation such as respiratory or cardiac depression.
medications, should be ingested as ordered, unless prescribed
by the invasive cardiologist. Pulmonary Ventilation
A preprocedure assessment will include an evaluation of One of the principle adverse effects from oversedation is drug-
anatomic variables that may affect sedation or, if necessary, induced respiratory depression. Therefore, the ASA strongly
intubation, and allow proactive planning to take place. In recommends monitoring respiratory ventilation via observation
planning for potential intubation, examination specific to the or auscultation during moderate sedation, despite little litera-
airway is recommended, including evaluation of the neck and ture to suggest this practice to be of value in preventing drug-
dentition. The Mallampati Scale (3), an objective anatomical induced respiratory effects. While there are some noninvasive
assessment of the oral cavity that predicts ease of intubation, is techniques to assess ventilation, such as impedance plethys-
also a useful examination tool (Table 6.3). In addition, a general mography, these are considered complementary to observation
physical assessment is also warranted. This should include and auscultation, and not substitutes.
[16/6/010/10:12:11] [47–52]
Table 6.1 American Society of Anesthesiology Continuum of Sedation

Minimal sedation
(anxiolysis) Moderate sedation/analgesia Deep sedation/analgesia General anesthesia
Responsiveness Normal response to Purposeful response to verbal Purposeful response following Unarousable even with
verbal stimulation or tactile stimulation repeated or painful painful stimulus
stimulation
Airway Unaffected No intervention required Intervention may be required Intervention often
required
Spontaneous Unaffected Adequate May be inadequate Frequently inadequate
ventilation
Cardiovascular Unaffected Usually maintained Usually maintained May be impaired
function
Table 6.2 Preprocedure Evaluation

System Rationale
Preexisting cardiac or pulmonary Sedative agents can cause cardiovascular or respiratory depression.
disease
Preexisting renal or hepatic disease Abnormalities may impair how fast the drug is metabolized and excreted from the system, resulting
in longer drug action and increased drug effect.
Time and type of last oral intake Reflex suppression could result in aspiration.
History of drug and alcohol abuse The dose and action of sedative agents may be affected in patients that abuse drugs and alcohol. In
addition, procedural agitation may be higher.
History of smoking Patients who smoke are at increased risk of bronchospasm, airway problems, or coughing.
Previous experience with sedative Any previous adverse reactions to sedation should be noted.
agents
Table 6.3 The Mallampati Scale Hemodynamics

Pharmacologic agents used in moderate sedation have the
Anatomic findings Difficulty of intubation
potential to blunt autonomic response to the procedure.
Class I Soft palate, uvula, fauces, No difficulty Hemodynamic variables such as blood pressure and heart
pillars available rate may indicate inadequate sedation when elevated. How-
Class II Soft palate, uvula, fauces No difficulty ever, sedation agents may also adversely depress these varia-
available
bles. Additionally, some local anesthetics and sedative agents
Class III Soft palate, base of uvula Moderate difficulty
available may cause cardiac dysrhythmias which may be more pro-
Class IV Hard palate only visible Severe difficulty nounced in patients with extensive cardiovascular disease. The
ASA recommends monitoring vital signs at five-minute inter-
vals until adequate sedation is achieved. However, given the
higher risk of the cardiovascular patient and the need for
Oxygenation general hemodynamic data for procedural effectiveness and/
Pulse oximetry has been shown to effectively monitor and or safety, this monitoring may occur at more frequent intervals
prevent hypoxemia during moderate sedation and should be (or continuously) during cardiac catheterization. In addition,
utilized in all patients undergoing cardiac catheterization. It continuous electrocardiographic monitoring is warranted to
should also be noted that pulse oximetry is not singularly monitor sedative adverse effects, although this is also gener-
adequate for assessing respiratory function and should be ally part of standard monitoring independent of sedation
combined with assessment tools for pulmonary ventilation. practices.
Table 6.4 The Aldrete Scoring System for Sedation

Scorea Activity Respiration Circulation Consciousness Color
2 Able to move four Able to breathe deeply BP 20% of Fully alert and answers Normal pink
extremities and cough baseline questions
1 Able to move two Limited respiratory effort BP 20–50% of Arousable Pale, dusky,
extremities (dyspnea) baseline blotchy
0 Not able to control any No spontaneous BP >50% of Failure to elicit response Frank cyanosis
extremities respiratory effort baseline
a
Scores of 8 or greater are generally considered acceptable to indicate recovery from sedation.
[16/6/010/10:12:11] [47–52]
CONSCIOUS SEDATION (LOCAL ANESTHETICS, SEDATIVES, AND REVERSING AGENTS) 49
Table 6.5 American Society of Anesthesiology–Recommended a-subunit of voltage-gated sodium channels in the nerve mem-
Emergency Equipment for Moderate Sedation brane. In addition, there are effects on G protein–coupled
receptors, calcium channels, and potassium channels which
Airway Compressed oxygen
complement the classically known effect on sodium channels
management Endotracheal tubes
equipment Face masks (5,6). Inhibition of these pathways essentially results in tempo-
Lubricant rary cessation of nerve impulse conduction.
Stylets In general, local anesthetics are both safe and effective
Suction although some significant pharmacologic differences exist
Suction catheters among agents (Table 6.6). Lidocaine (0.5–2% concentration) is
Intravenous Alcohol wipes the local anesthetic of choice in cardiac catheterization due to
equipment Gloves its rapid onset, short duration of action, and minimal risk of
Intravenous catheters cardiotoxicity. Cardiotoxicity induced by local anesthetics can
Intravenous fluids
be caused by several mechanisms, but classically is described as
Intravenous tubing
both direct and indirect (central nervous system mediated)
Needles for drug administration or
intramuscular or intraosseous injection effects on the myocardium with conduction delays leading to
Sterile gauze pads a prolonged PR interval or a wide QRS complex (8). In addition,
Syringes many anesthetic agents can cause ventricular arrhythmias
Tape through this same mechanism by unidirectional block and
Tourniquets reentry (9,10), the risk of which may be heightened by mechan-
Medications Amiodarone ical disturbance of myocyte electrophysiology during cardiac
Atropine catheterization. Some agents also possess strong negative ino-
Dextrose (50%) tropic activity which is generally not a desirable pharmacologic
Diazepam or midazolam
property for administration to a patient with significant struc-
Diphenhydramine
tural heart disease (11). Bupivacaine is widely regarded as the
Epinephrine
Hydrocortisone, methylprednisolone, or local anesthetic with the highest potential for cardiotoxicity and
dexamethasone should not be considered for use in cardiac catheterization (8).
Lidocaine Since most anesthetics are vasodilators, a vasoconstrictor
Nitroglycerin (typically epinephrine at 1:100,000 or 1:200,000) may be utilized
Vasopressin in combination with a local anesthetic to minimize local bleed-
ing and improve anesthetic duration. Use of a vasoconstrictor
Source: From Ref. 2. has not been shown to significantly affect hemodynamic param-
eters (12,13), although little data exist to establish the safety of
local vasoconstrictor use in the patient with high cardiovascular
EQUIPMENT acuity. Of note, the administration of a vasoconstrictor in com-
The ASA recommends a minimum level of emergency equip- bination with a local anesthetic will reduce the potency of the
ment in the event of pulmonary or cardiac arrest induced by anesthetic employed by reducing tissue diffusion, necessitating
procedural sedation and/or analgesia (Table 6.5). Also, given higher doses for equivalent effect. This also lowers the risk of
the high cardiovascular acuity of the patient undergoing car- local anesthetic toxicity by minimizing vascular uptake via
diac catheterization and the potential for cardiac and/or pul- vasoconstriction of capillary beds (14,15).
monary arrest independent of sedation use, the ACC/SCAI
recommends a baseline level of emergency resuscitative equip-
ment be immediately available (1). In addition, cardiac cathe- Sedation and Analgesia
terization laboratory personnel should possess a minimum Cardiac catheterization is most often performed under the influ-
level of training (basic life support) and advanced levels of ence of agents with anxiolytic and amnestic properties in com-
training are highly recommended. bination with an analgesic, if necessary. The most frequently
used sedative agent is a benzodiazepine, which may also be
combined with a sedating antihistamine. Opioid analgesics may
CLINICAL ASPECTS also be utilized in combination with benzodiazepines to achieve
Local Anesthesia adequate sedation. Literature suggests that combining a seda-
Local anesthesia continues to be the preferred method to relieve tive with an opioid provides adequate moderate sedation (16),
pain and sensation associated with vascular access. Local but there are few data describing the superiority of the combi-
anesthetics primarily act by competitive antagonism of the nation versus either agent alone. Furthermore, data suggest that
Table 6.6 Selected Local Anesthetic Agents

Maximum dose with
Agent Onset Potency Duration Cardiotoxicity Maximum dose (mg/kg) vasoconstrictor (mg/kg)
Lidocaine þþþ þþ þþ þ 5 7
Mepivicaine þ þþ þþ þ 5 7
Bupivacaine þ þþþ þþþ þþþ 2 3
Ropivacaine þ þþ þþþ þþ 2.5 4
Note: þ, low; þþ, intermediate; þþþ, high.
[16/6/010/10:12:11] [47–52]
Table 6.7 Selected Benzodiazepines for Use in Procedural Sedation

Duration Metabolic Active Protein
Dose Peak effect of effect pathway metabolite? binding (%)
Midazolam IV: 0.5–2 mg 3–5 min 30–80 min Oxidation Yes 95
Maximum: 5–10 mg
Diazepam Oral: IV: Oral: 30 min 2–4 hr Oxidation Yes 80
5 mg 5–10 mg IV: 8–10 min
Maximum: 20 mg
Lorazepam Oral: IV: Oral: 60–90 min 6–8 hr Conjugation No 85
4 mg 2 mg IV: 15–20 min
Maximum: 4 mg
the combination of a sedative agent with an opioid increases the system and produce hypnosis and anxiolysis, in addition to
risk of adverse outcomes such as respiratory depression and an amnestic effect. Several methods of dosing are commonly
hypoxemia (17). Therefore, the combination of a benzodiazepine utilized in the cardiac catheterization laboratory, including
with an opioid may be utilized, but ideally, therapy should be premedication with oral benzodiazepines (such as diazepam)
individualized for each specific patient’s needs and adminis- and/or the use of rapid and short acting benzodiazepines (such
tered separately to determine the response of the patient and if as midazolam). Drug properties of each benzodiazepine are
the desired level of sedation can be achieved with minimal use listed in Table 6.7. Significant differences exist among agents in
of drug stacking. This also underscores the need for continuous terms of onset and duration of action and institutional practice
monitoring of the patient not only to achieve adequate sedation, will likely be tailored for the specific agent preferred. There are
but also to prevent adverse outcomes associated with the effort. no data sustaining the superiority of any benzodiazepine in
Repeat dosing may be required with longer procedures. terms of efficacy or safety for moderate sedation. Patients with
altered clearance or significant drug-drug interactions will
Antihistamines require either alternative benzodiazepines or lower doses. For
Oral antihistamines, especially diphenhydramine, have been example, patients who are elderly or those with significant liver
described as early as the 1960s as sedative medication for disease will have prolonged clearance of diazepam due to
cardiac catheterization (18). They remain a popular preproce- slower oxidative activity within the liver and may have shorter
dural medication, typically in combination with an oral benzo- recovery time with a drug that is conjugated without active
diazepine, because of the sedative properties and also to blunt metabolite formation (e.g., lorazepam).
unreported or unknown allergic reactions to contrast media. Adverse effects of benzodiazepines in moderate sedation
First-generation antihistamines exert sedative effects largely generally relate to respiratory depression as a result of over-
through a combination of anticholinergic properties and lip- sedation. Hemodynamic effects are minimal, but hypotension
ophilicity, allowing the drug to easily cross the blood-brain can result with rapid administration of intravenous benzodiaze-
barrier and affect histamine mediated neurologic function (19). pines containing propylene glycol (e.g., diazepam, lorazepam).
While diphenhydramine is most often used as a proactive
baseline level of preprocedure sedation, additional reactive Opiates
methods may be required on the basis of the individual patient. The opiates, such as morphine, fentanyl, and meperidine, are
Of note, in part because of the strong anticholinergic effects commonly used for analgesia in moderate sedation practice in
of diphenhydramine, elderly individuals appear to have more combination with local anesthesia. Opiates are preferred not
pronounced and prolonged sedative effects (20). In combination only because of their superior pharmacologic profiles for pro-
with diazepam, a drug that is also slowly metabolized by elderly cedural analgesia (quick onset and quick offset) but also
patients, this may result in excessive and/or prolonged sedation because of their lack of interaction with the renal prostaglandin
and may be considered a less preferred method of achieving system (unlike nonsteroidal compounds), resulting in safer
adequate baseline procedural sedation for all patients. interactions with nephrotoxic contrast media. Unfortunately,
the opiates carry a greater risk of respiratory depression, par-
Benzodiazepines ticularly in combination with benzodiazepines. Caution and
The most commonly used agents for sedation in cardiac judicial use should be applied to their use, along with careful
catheterization are the benzodiazepines, which act as the g- monitoring. Table 6.8 lists the various opiates that could be
aminobutyric acid (GABA) complex in the central nervous considered for use during cardiac catheterization. Of note,
Table 6.8 Selected Opiates for Use in Procedural Sedation

Duration Active Cardiovascular Protein binding
Dose Peak effect of effect metabolite? effects (%)
Fentanyl IV: 25–50 mcg 1 min 30–60 min No Low 80–86
Maximum: 3 mcg/kg
Morphine IV: 1–2 mg 1–2 min 3–4 hr Yes Moderate 20–30
Maximum: 0.15 mg/kg
Meperidine IV: 10–20 mg 1–2 min 2–3 hr Yes Low 65–80
Maximum: 1.5 mg/kg
[16/6/010/10:12:11] [47–52]
CONSCIOUS SEDATION (LOCAL ANESTHETICS, SEDATIVES, AND REVERSING AGENTS) 51
Table 6.9 Pharmacologic Reversal of Sedation and Analgesia

Dose Peak effect Duration of effect
Flumazenil (benzodiazepine antagonist) 0.2 mg initial 6–10 min 1–4 hr
Maximum (total): 1 mg
Naloxone (opioid antagonist) 0.4 mg initial 5–10 min 45 min–3 hr
Maximum (total): 2 mg
meperidine should not be used in patients with severe renal FUTURE DIRECTIONS
insufficiency as seizures may be provoked via active metabolite While modern day methods to achieve moderate sedation are
accumulation, and should also be considered less preferred in generally adequate, room remains for improvement in the
general because of the risk of contrast nephropathy with car- efficacy and particularly the safety of sedation practice. An
diac catheterization. ideal agent would be a drug that has rapid onset and offset,
minimal effects on respiratory function at maximal efficacy,
and is not affected by patients with altered organ function.
SPECIAL ISSUES However, widespread inclusion of a new agent into practice
Local Anesthetic Allergies may be cost prohibitive, and benefit to the patient must be
While patients occasionally report a history of an allergic reac- demonstrated.
tion to a local anesthetic, true IgE-mediated hypersensitivity Some literature exists for the use of ketamine as an
reaction has not been not documented in the literature (21). In alternative analgesic agent which also produces a “dissocia-
many cases, such reactions are misclassified adverse effects or tive” anesthetic effect (25). Although the mechanism of action
allergic responses from preservatives (methylparaben) or other of ketamine is not entirely clear, it appears to act on a wide
excipients contained within the anesthetic solution, such as variety of central nervous system receptors, including
bisulfate compounds designed to prevent oxidation of vaso- N-methyl-aspartate (NMA), opiate, serotonin, and norepi-
constrictors. Certain local anesthetics containing ester com- nephrine receptors (25). One clear advantage of ketamine in
pounds (prilocaine, tetracaine, or benzocaine), which are procedural sedation is that it has only minimal effects on
derivatives of para-aminobenzoic acid (PABA), may cause an respiratory function. This may make ketamine more desirable
allergic reaction since their structures contain a potentially in situations where intensive analgesia is required.
immunogenic amine substitution. When confronted with a Dexmedetomidine has also gained considerable interest
history consistent with an anesthetic induced allergic reaction, as a central a-2 agonist that achieves adequate sedation with-
the safest course of action without undergoing allergy testing out affecting respiratory function (26). Although dexmedeto-
would be to choose an amide anesthetic (lidocaine, mepivi- midine is more selective for a-2 receptors than its structural
caine, bupivacaine, or ropivacaine) in a preservative free solu- relative clonidine (26), significant decreases in blood pressure
tion [typically denoted as “methylparaben free” (MPF)] that and heart rate have been noted with its use that may limit
does not contain a vasoconstrictor. widespread applicability to patients with significant cardiovas-
cular disease.
Reversal of Sedation and Analgesia

Pharmacologic antagonists are available for both opiates and CONCLUSIONS
benzodiazepines (Table 6.9) and should be immediately acces- The preferred level of sedation during cardiac catheterization is
sible for use. Naloxone is a competitive antagonist for all defined as moderate sedation. However, sedation level is
receptors affected by opiates (m, w, s) and it possesses no highly continuous and requires constant monitoring to ensure
agonist effects. Complete antagonism of analgesic response, adequate efficacy without oversedation. Short acting benzodia-
although sometimes necessary, can result in severe onset of zepines are the preferred pharmacologic agents to induce seda-
pain and physiologic pain response (hypertension, tachycar- tion in the catheterization laboratory and may be paired with
dia) and may require further intervention. Flumazenil is an opioid analgesia, however, extreme care must be taken to avoid
antagonist of the benzodiazepine receptor and has been shown oversedation with this combination. While tremendous vari-
to be beneficial in reversing the central nervous system ability in individual sedation practice exists, a systematic
depressant effects of the benzodiazepines. Reversal of benzo- approach to using conscious sedation will aid in improving
diazepine induced respiratory depression is less conceptually patient comfort during cardiac catheterization while also avoid-
clear, although literature does exist supporting its role in this ing potential adverse effects associated with current sedation
area (22,23). Caution must also be exercised in complete practices.
reversal in the patient receiving prior long-term benzodiaze-
pine therapy, as this may provoke withdrawal symptoms,
including seizures (24). Of note, many benzodiazepines and
opiates have longer elimination half-lives than their pharma-
REFERENCES
1. Gross JB, Bailey PL, Connis RT, et al. Practice guidelines for
cologic antagonists. Successful reversal should be followed up sedation and analgesia by non-anesthesiologists. Anesthesiology
with an appropriate duration of monitoring based on the 2002; 96:1004–1017.
elimination half-life of the offending agent and particularly 2. American Society of Anesthesiology. ASA Standards, Guidelines,
close attention should be given to patients with altered drug and Statement. Park Ridge: American Society of Anesthesiology,
clearance. 2006.
[16/6/010/10:12:11] [47–52]
3. Mallampati SR, Gatt SP, Gugino LD, et al. A clinical sign to predict 14. Karmakar MK, Ho AM, Law BK, et al. Arterial and venous phar-
difficult tracheal intubation: a prospective study. Can Anaesth Soc macokinetics of ropivacaine with and without epinephrine after
J 1985; 32:429–434. thoracic paravertebral block. Anesthesiology 2005; 103:704–711.
4. Aldrete JA, Kroulik D. A postanesthetic recovery score. Anesth 15. Yagiela JA. Intravascular lidocaine toxicity: influence of epinephr-
Analg 1970; 49:924–934. ine and route of administration. Anesth Prog 1985; 32:57–61.
5. Hollmann MW, Wieczorek KS, Berger A, et al. Local anesthetic 16. Kennedy PT, Kelly IM, Loan WC, et al. Conscious sedation and
inhibition of G protein-coupled receptor signaling by interference analgesia for routine aortofemoral arteriography: a prospective
with Galpha(q) protein function. Mol Pharmacol 2001; 59:294–301. evaluation. Radiology 2000; 216:660–664.
6. Xiong Z, Strichartz GR. Inhibition by local anesthetics of Ca2þ 17. Bailey PL, Pace NL, Ashburn MA, et al. Frequent hypoxemia and
channels in rat anterior pituitary cells. Eur J Pharmacol 1998; apnea after sedation with midazolam and fentanyl. Anesthesiol-
363:81–90. ogy 1990; 73:826–830.
7. Dunn SP. Local anesthetics, sedatives, and reversing agents. In: 18. Akdikmen SA, Rosenthal R, Landmesser CM. Diphenhydramine
Moliterno DJ, Mukherjee D, Popma JJ, et al., eds. CathSAP 3 hydrochloride in premedication for cardiac catheterization.
(Cardiac Catheterization and Interventional Cardiology Self- Anesth Analg 1966; 45:293–297.
Assessment Program). Washington, DC: American College of 19. Carruthers SG, Shoeman DW, Hignite CE, et al. Correlation
Cardiology Foundation, 2008:437–443. between plasma diphenhydramine level and sedative and anti-
8. Mather LE, Chang DH. Cardiotoxicity with modern local anaes- histamine effects. Clin Pharmacol Ther 1978; 23:375–382.
thetics: is there a safer choice? Drugs 2001; 61:333–342. 20. Agostini JV, Leo-Summers LS, Inouye SK. Cognitive and other
9. Klein SM, Pierce T, Rubin Y, et al. Successful resuscitation after adverse effects of diphenhydramine use in hospitalized older
ropivacaine-induced ventricular fibrillation. Anesth Analg 2003; patients. Arch Intern Med 2001; 161:2091–2097.
97:901–903. 21. Becker DE, Reed KL. Essentials of local anesthetic pharmacology.
10. Lipka LJ, Jiang M, Tseng GN. Differential effects of bupivacaine Anesth Prog 2006; 53:98–108.
on cardiac K channels: role of channel inactivation and subunit 22. Davis DP, Hamilton RS, Webster TH. Reversal of midazolam-
composition in drug-channel interaction. J Cardiovasc Electro- induced laryngospasm with flumazenil. Ann Emerg Med 1998;
physiol 1998; 9:727–742. 32:263–265.
11. David JS, Ferreti C, Amour J, et al. Effects of bupivacaine, 23. Pitman V, Stoll P, Hoffman J. Flumazenil for reversal of respira-
levobupivacaine and ropivacaine on myocardial relaxation. Can tory depression. Am J Hosp Pharm 1994; 51:235–236, 239.
J Anaesth 2007; 54:208–217. 24. McDuffee AT, Tobias JD. Seizure after flumazenil administration
12. Niwa H, Satoh Y, Matsuura H. Cardiovascular responses to in a pediatric patient. Pediatr Emerg Care 1995; 11:186–187.
epinephrine-containing local anesthetics for dental use: a compar- 25. White PF, Way WL, Trevor AJ. Ketamine–its pharmacology and
ison of hemodynamic responses to infiltration anesthesia and therapeutic uses. Anesthesiology 1982; 56:119–136.
ergometer-stress testing. Oral Surg Oral Med Oral Pathol Oral 26. Peden CJ, Prys-Roberts C. Dexmedetomidine–a powerful new
Radiol Endod 2000; 90:171–181. adjunct to anaesthesia? Br J Anaesth 1992; 68:123–125.
13. Rosenberg M. The cardiovascular response of oral and maxillofa-
cial surgeons during administration of local and general anesthe-
sia. J Oral Maxillofac Surg 1987; 45:306–308.
[2/7/010/7:23:1] [53–66]
Vasopressors, vasodilators, and antithrombotics

in the catheterization laboratory
Tracy E. Macaulay and David J. Moliterno
INTRODUCTION output (5). Use of inotropic agents such as dobutamine and

This chapter provides an overview of medications commonly milrinone to achieve an adequate cardiac index of 2.2 L/min/m2
used during cardiac catheterization and interventional proce- may provide necessary support. However these agents have not
dures. It will focus on medications that provide hemodynamic been shown to provide a mortality benefit (6–8). Temporary
optimization during procedures as well as agents used to increases in cardiac preload and afterload may also be required
prevent thrombotic complications. Other chapters addressing and can be achieved by increasing volume (blood or intravenous
specific therapeutic areas (e.g., ST segment elevation myocar- fluids) and/or stimulation of a-receptors in the periphery.
dial infarction) may address individual treatments in more When treating patients with vasopressor therapy, target-
detail, while this chapter will serve as a broad overview of ing a minimal perfusion pressure or mean arterial pressure of
pharmacology, therapeutic applications, monitoring, and safety approximately 65 mmHg is widely accepted (9). However,
of these agents. because blood pressure alone is not a determinant of adequate
oxygen delivery, advanced hemodynamic monitoring and indi-
vidualized goals are appropriate. This will allow for selection of
VASOACTIVE OVERVIEW the appropriate vasoactive medication based on the clinical
In critically ill patients, hemodynamic stability can be achieved situation and limit unnecessary exposure or deleterious side
using mechanical support with intra-aortic balloon pumps and effects.
assist devices, however an initial pharmacologic approach is
typically preferred. The main goal of using vasoactive medi-
cations is to maintain tissue perfusion and optimize oxygen PHARMACOLOGIC FUNDAMENTALS
delivery. Although the hemodynamic goals vary depending on Vasodilators
the situation, the overall goal of meeting metabolic demands NO
and preventing multisystem organ dysfunction and death Nitric oxide (NO) is a potent vasodilator (Table 7.1) that is
remains the same (1,2). When considering the use of vaso- decreased in patients with coronary artery disease. Nitroglyc-
dilators, inotropes, and vasopressors keep in mind that the aim erin provides an exogenous source of NO, which works by
is to provide adequate oxygen delivery while maintaining a increasing vascular cyclic guanosine monophosphate (cGMP)
minimal effective perfusion pressure. This approach limits the resulting in smooth muscle relaxation. The majority of nitro-
myocardial oxygen demand and ischemia, as well as other glycerin’s effect is in preload reduction making it an ideal
deleterious effects of using these agents. choice for the patient with hypertension and elevated pulmo-
nary capillary wedge pressure. For the same reason caution
should be used in patients that are preload dependent, such as
ANATOMIC CONSIDERATIONS WITH those with right ventricular failure. Nitroglycerin also causes
VASOACTIVE AGENTS endothelium-independent coronary artery dilation, antago-
Vasodilator therapy is most frequently used in the treatment of nizes vasoconstriction and vasospasm and increases collateral
hypertension. It is also useful in patients with left ventricular blood flow. These effects make it useful in treatment of ische-
systolic dysfunction (LVSD) with or without pulmonary mia symptoms, although no mortality reduction has been
edema, and has demonstrated a mortality reduction in both demonstrated in large clinical trials (11,12). Tachyphylaxis
acute and chronic management (3,4). In the catheterization can develop within 24 hours of continuous therapy, but can
laboratory these medications can be given systemically to generally be overcome by increasing the dose or providing a
lower peripheral vascular resistance or locally as direct coro- nitrate-free interval if possible (5).
nary arterial vasodilators. Some vasodilators may also be used
in the characterization and treatment of pulmonary arterial Sodium Nitroprusside
hypertension. Sodium nitroprusside is an endothelium-independent nitrovaso-
Intravenous inotropic therapy is often necessary for the dilator like nitroglycerin, however it has greater arterial vaso-
treatment of cardiogenic shock, particularly in patients with dilating properties than venous (13). Unlike other vasodilators,
LVSD. In cardiogenic shock, a decrease in cardiac output nitroprusside causes only mild increases in heart rate and an
results in a hypoperfusion state, thereby increasing adrenergic overall decrease in myocardial oxygen demand. The molecular
drive. The release of endogenous catecholamines is temporarily composition of sodium nitroprusside includes five cyanide ions
effective, however it may be necessary to stimulate b1-receptors which can accumulate as either cyanide or thiocyanate in
or administer phosphodiesterase inhibitors to augment cardiac patients with liver or renal failure, respectively. Patients
[2/7/010/7:23:1] [53–66]
Table 7.1 Vasodilator Overview

Maintenance Intracoronary Precautions
Generic (brand) Starting dose and titration dose ADR and CI Uses
Nitroglycerin 5 mcg/min Increase every 50–250 mcg Headache, RV failure Used to treat symptoms
(nitrostat IV) 3 min to tachyphylaxis (preload of ischemia at low
maximum dependent doses (10),
200 mcg/min patients) hypertension with or
without pulmonary
edema at higher
doses
Nitroprusside 0.5 mcg/kg/min Increase every 25–100 mcg With prolonged Severe renal Cardiac failure due to
(nipride) 5 min to use (or high or liver increase afterload,
maximum of doses) cyanide impairment, ADHF, vasodilatory
5–10 mcg/kg/ toxicity and COPD challenges
min (ideally methemo- (pulmonary HTN)
for <48 hr) globinemia
may develop
Diltiazem For arrhythmias 5–20 mg/hr AV blockade LVSD, AV AF/flutter, PSVT
(cardizem) 10 mg IVP continuous blockade,
(0.25 mg/kg) infusion. sick sinus
Consider syndrome,
rebolus with WPW
each
uptitration.
Verapamil For arrhythmias 100–200 mcg AV blockade LVSD, AV AF/flutter, PSVT
(isoptin) 2.5–5 mg IVP (up to blockade,
4 times) sick sinus
syndrome,
WPW
Nicardipine 3–15 mg/hr Hypotension, Severe aortic Stability of more
(cardene) tachycardia, stenosis, concentrated solution
headache, peripheral unknown, therefore
peripheral edema infusion provides
edema large volume of fluid
Clevedipine Hypotension, Hypertension
(cleviprex) tachycardia,
headache,
peripheral
edema
Adenosine 12–24 mcg
Abbreviations: ADR, adverse drug reaction; AV, atrioventricular; LVSD, left ventricular systolic dysfunction.
Table 7.2 Comparison of Intravenous Vasodilators

NTG Nitroprusside Diltiazem Verapamil Nicardipine Clevedipine
MOA Exogenous NO,
:cyclic guanosine
monophosphate
Overall Venovasodilation Arterial Decrease Decrease Decrease Decrease
hemodynamic and venous heart rate and heart rate and blood blood
effect dilator vasodilaton vasodilation pressure pressure
; Contractility 0 0 2 4 0
: Heart rate 0 1 (reflex) 0/1 0/1 1 (reflex)
; SA automaticity 0 0 4 4 0
; Atrioventricular 0 0 4 5 0
conduction
Vasodilation 2 4 3 4 5
receiving high doses (10 mcg/kg/min) for a prolonged time Adenosine

(>2 days) are particularly at risk of toxicity. Despite disadvan- Adenosine is a purine nucleoside in human cells involved in
tageous toxicities, nitroprusside is the gold standard intrave- cellular metabolism. High doses of exogenous adenosine slow
nous antihypertensive for its remarkable effectiveness and sinus node conduction interrupting reentry pathways and
rapid onset of action (Table 7.2). restoring normal sinus rhythm (14,15). The most common
[2/7/010/7:23:1] [53–66]
VASOPRESSORS, VASODILATORS, AND ANTITHROMBOTICS IN THE CATHETERIZATION LABORATORY 55
therapeutic use for adenosine is treatment of atrioventricular Dobutamine and Isoproterenol

(AV) nodal reentrant tachycardia, which is terminated with a Dobutamine and isoproterenol primarily act as a b1-receptor
near 90% success rate. For this purpose adenosine is given as a agonists, but dobutamine also exhibits mild b2- and a-receptor
6 mg IV rapid injection. After two minutes 12 mg can be given effects (b1 >> b2 > a) (16). For inotropy in left ventricular heart
if needed, and then repeated for a total dose of 30 mg. When failure dobutamine doses as low as 1 mcg/kg/min and up to
administering adenosine, it is important to follow each dose 20 mcg/kg/min have been used. Niches in therapy for iso-
with an intravenous flush to ensure rapid distribution, as the proterenol include treatment of bradycardia post-cardiac trans-
onset is immediate and the duration of effect is mere seconds. plantation and to prevent reoccurrence of torsades de pointes
When given through a central line these doses should be (17). With minimal a-receptor activity dobutamine and isopro-
halved, as there are case reports of adenosine initiating atrial terenol may be administered through a peripheral vein; how-
fibrillation. In cardiac catheterization, adenosine can be used to ever, they would not be indicated as monotherapy in shock as
evaluate coronary flow reserve since it causes maximal coro- they provide little effect on blood pressure.
nary vasodilation. The greatest benefit of the short duration is
that any adverse effects (angina, dyspnea, AV nodal block, and Milrinone
flushing) are quickly self-limiting. Milrinone is a selective phosphodiesterase inhibitor that
increases intracellular cAMP in cardiac tissue and vascular
Calcium Channel Blockers PDE III (8). Pharmacologic effects include positive inotropy
Calcium channel blockers (CCB) are a large and growing class of (increasing cardiac output) and peripheral vasodilation
medications. The unifying mechanism is inhibition of calcium (decreasing cardiac demand and PCWP). Additional beneficial
from entering “slow channels” or voltage-sensitive areas of effects include both positive lusitropy and pulmonary arterial
vascular smooth muscle, producing smooth muscle relaxation vasodilation. Like dobutamine, milrinone is indicated for
and vasodilation. CCB are further classified as dihydropyridine severe left ventricular dysfunction, therefore differences in
and nondihydropyridine on the basis of differences in structure these drugs’ effects and pharmacokinetic profiles help guide
and pharmacologic effects. Diltiazem and verapamil are non- therapy choices. Milrinone may be more beneficial than dobut-
dihydropyridines and offer greater negative chronotropic amine in patients on chronic b-blockade therapy based on
effects than dihydropyridines. Both agents slow AV nodal nonadrenergic mechanism and in those with normal to high
conduction and exert a negative inotropic effect with less blood pressures. However, milrinone is renally eliminated and
arterial vasodilation. In the catheterization laboratory these should be avoided or administered at a reduced dose in
agents are most often used for treatment of rapid ventricular patients with severe renal dysfunction. No difference in clinical
heart rate secondary to atrial fibrillation. Nifedipine is the outcomes or arrythmogenicity has been demonstrated in com-
classic dihydropyridine CCB, although newer second-genera- parisons of dobutamine and milrinone. Of note, short-term use
tion agents are more often used in an acute setting. Dihydro- of milrinone for mild to moderate heart failure exacerbations
pyridines have selectivity for peripheral vasculature, resulting has not been shown to decrease hospital length of stay and has
in extensive blood pressure lowering with little to no effect on been associated with an increase in arrhythmias (7), further
heart rate. In fact, more often compensatory increases in heart supporting that inotropic agents should be reserved for severe
rate are seen. Use in the setting of acute coronary syndrome heart failure.
(ACS) and in patients with heart failure is controversial, as they Vasopressors During cardiac catheterization it may be
can result in fluid retention in addition to adrenergic activation. necessary to provide temporary hemodynamic support for a
Inotropes Inotropic agents are indicated for the treat- patient in shock or experiencing extremely low blood pressure.
ment of low-output failure in patients with elevated left ven- This sometimes requires use of vasopressors, most commonly
tricular filling pressures (15). Effects are either directly dopamine, norepinephrine, epinephrine, and phenyleprine
mediated through b1-receptor agonism or indirectly through (Table 7.4). Unlike the inotropic and vasodilatory medications
an increase in intracellular cAMP, raising heart rate and there- previously discussed, administration of medications with vaso-
fore cardiac output (Table 7.3). All inotropes have arrhythmo- constrictive properties requires use of a central line to avoid
genic properties and therefore should be reserved for acutely ill limb ischemia that can occur with extravasation. If extravasa-
patients and used for the shortest duration necessary. tion does occur, phentolamine, an a-adrenergic blocking agent,
Table 7.3 Inotrope Overview

Generic Brand Maintenance
name name Starting dose and titration PK properties Adverse effects
Dobutamine Dobutrex 1 mcg/kg/min 20 mcg/kg/min Metabolized by methylation Angina, hypertension,
and conjugation, inactive tachyarrhythmia,
metabolites are renally headache
excreted, t1/2 ¼ 2 min
Isoproterenol Isuprel 0.01 mcg/kg/min Increase every Hepatic metabolism, Syncope, tachyarrhythmia,
5 min to maximum t1/2 ¼ 3–7 hr confusion, tremor
0.3 mcg/kg/min
Milrinone Primacor 0.25 mcg/kg/min 0.75 mcg/kg/min Renally excreted (83% Ventricular arrhythmias,
unchanged), t1/2 ¼ 2.3 hr hypotension, headache
(prolonged in renal failure)
Abbreviation: PK, pharmacokinetic.
[2/7/010/7:23:1] [53–66]
Table 7.4 Vasopressor Overview

Drug type and generic name Brand name Starting dose Maintenance and titration
Dopamine Intropin 3–10 mcg/kg/min Up to 20 mcg/kg/min
Epinephrine Adrenalin 1 mcg/min 10 mcg/min
Norepinephrine Lovephed 2 mcg/min 30 mcg/min
Phenylephrine Neo-Synephrine 100 mcg/min Decrease to maintenance of 40–60 mcg/min
Vasopressin Pitressin 0.01 units/min 0.04 units/min
can be injected directly into the areas of ischemic tissue, and can be initiated or until the patient’s hemodynamics have been
warm compresses should be applied. adequately stabilized. To quickly make a 10 mcg/mL syringe of
epinephrine, one can use 0.1 mL of 1:1000 epinephrine or (more
Dopamine easily) 1 mL of 1:10,000 epinephrine and dilute this with saline
Dopamine exhibits unique pharmacologic actions according to for a total of 10 mL. For less severe reactions, epinephrine 0.3 to
the dosage. Low doses (<5 mcg/kg/min) have primarily dop- 0.5 mg (0.3–0.5 mL of 1:1000 solution) can be given subcuta-
aminergic receptor (DA1) activity, resulting in vasodilation neously or intramuscularly. Repeated doses are often necessary
increasing renal, mesenteric, and coronary blood flow. Histor- while steroids and other symptomatic treatments take effect.
ically low or “renal-dose” dopamine was thought to be renal Vasopressin can be used as an alternative to epinephrine
protective and could aid in prevention of contrast-induced in acute cardiac arrest. In these situations vasopressin is given
nephropathy. However, study results have been conflicting as a bolus of 40 units (22). The proposed mechanism of vaso-
and overall show no evidence of a true renal-protective benefit pressin activity is maintenance of vascular tone and modula-
or clinical outcome improvement with renal-dose dopamine tion of cardiovascular homeostatis (21,23). One potential
(18–20). Dopamine doses of 3 to 10 mcg/kg/min will stimulate advantage of the noncatecholamine, vasopressin, over epi-
b1- and b2-receptors (b1 > b2), resulting in similar inotropic nephrine is in patients with acidosis, as catecholamine response
properties to dobutamine. At maximal doses of 10 to 20 mcg/ is diminished in patients with low pH. Fixed-dose vasopressin
kg/min, a-receptor agonism overrides inotropy providing infusions are recommended as add-on therapy to norepineh-
blood pressure support. Doses in this range can increase car- prine in patients with vasodilatory shock.
diac demand, resulting in increased myocyte death and myo-
cardial ischemia, limiting its attractiveness in ACS. Pulmonary Phenylephrine
vein vasoconstriction can occur in patients receiving dopamine, Phenylephrine is the drug of choice when a clinical situation calls
therefore, using PCWP measurements as an estimate of left for pure vasoconstriction. Phenylephrine is an a-receptor ago-
ventricular end-diastolic filling pressures is less reliable (9). nist with no b-receptor activity. It is ideal for treatment of
anesthetic-induced hypotension or in the presence of unop-
Norepinephrine and Epinephrine posed parasympathetic activity (i.e., spinal cord injuries). Dos-
Norepinephrine and epinephrine are other vasoactive medications ing typically begins at 100 to 180 mcg/min until blood
with mixed receptor activity. Norepinephrine has b1- and a- pressures are stable and then infusion is decreased to a mainte-
receptor activity, with less b2. This translates clinically into nance dose of 40 to 60 mcg/min. The potent vasoconstrictor
more potent blood pressure affecting properties and less ino- activity may cause a reflexive decrease in adrenergic drive.
tropic effects at all dosage ranges as compared with dopamine. Therefore, caution should be used in patients with myocardial
For this reason norepinephrine remains the initial catechol- ischemia or heart failure induced shock, where inotropy is
amine of choice for septic or vasodilatory shock. Epinephrine is needed. Table 7.5 details the hemodynamic effects of vasopres-
the most potent agonist for the b1, b2 and a-receptors. Use of sors and inotropes.
continuous epinephrine infusion is limited by adverse effects,
including arrhythmias, ischemia, tachycardia, hyperglycemia,
cerebral hemorrhage, pulmonary edema, and diminished ANTITHROMBOTIC OVERVIEW
splanchnic blood flow. As such, epinephrine should be To prevent the two most frequent complications from percuta-
reserved for patients who are unresponsive to dopamine or neous coronary intervention (PCI), bleeding and thrombosis, it
norepinephrine (21). is important to achieve adequate protection from thrombosis
Epinephrine bolus dosing has a role in treatment of without resultant hemorrhage. This balance is reached by using
pulseless ventricular tachycardia, ventricular fibrillation, asys- an optimal combination of anticoagulant and antiplatelet med-
tole, or pulseless electrical activity. Epinephrine at doses of 0.5 to ications. The ideal balance would be the lowest possible yet
1 mg intravenously (0.1–0.5 mg intracardiac or 0.1 mg/kg given adequately effective regimen. This section examines current
via endotracheal tube) remains the first-line medication for use evidence for the efficacy and safety of antiplatelet and antith-
in advanced cardiac life saving and the most common reason for rombotic agents in PCI.
its use in catheterization procedures. Another emergent use is in
anaphylactoid reactions to medications, or more commonly
contrast. In this setting, depending on the degree of compromise, ANATOMIC CONSIDERATIONS WITH
different doses and routes of epinephrine administration may be ANTITHROMBOTIC AGENTS
appropriate. For systemic anaphylactoid reactions (i.e., hypo- Under normal circumstances the body’s hemostatic system is
tension), epinephrine should be given emergently as an intrave- finely regulated by opposing mechanisms. Prostacyclin, nitric
nous bolus dose of 10 mcg. This dose should be repeated in one- oxide, tissue plasminogen activator, thrombomodulin, protein
minute intervals until an intravenous infusion of epinephrine C, and protein S protect against coagulation in individuals with
[2/7/010/7:23:1] [53–66]
Table 7.5 Hemodynamic Effects of Vasopressors and Inotropesa

Drug type, generic (brand) HR MAP PAWP SVRI CO Receptor activity
Vasopressors
Dopamine (Intropin1) 2
<3 0 0 0 0 D > b1
3–10 1 1 0 1 b1 > b2 > D
>10 2 2 1 1 a1 > b1 >> b2
Epinephrine (Adrenalin1) 3
<0.05 0 0 0/1 3 b1 >> b2
0.05–0.15 1 2 1 3 b1 > a1 > b2
>0.15 2 2 3 2 a1 > b1
Norepinephrine (Lovephed1) 2 3 3 3.5 0 a1 >> b1
Phenylephrine (Neo-Synephrine1) 0 2 2 1 0 a1
Vasopressin
Inotropes
Dobutamine (Dobutrex1) 1 b 1 > b 2> a 1
2–10 1 1 1 1
10–20 0/1 1 0/1 2
Isoproterenol (Isuprel1) 2 2 3 3 b1 >> b2
Amrinone (Inocor1) 1 0 1 2 1 Phosphodiesterase inhibitor
Milrinone 1 2 1 2 2 Phosphodiesterase inhibitor
a
All listed doses are in mcg/kg/min.
intact endothelium. In cases of vascular damage, life-threat- TXA2-mediated platelet aggregation and vasoconstriction are
ening hemorrhage may be prevented by thrombus formation. inhibited for the life of the platelet (25). Aspirin has been shown
However, the same thrombotic mechanisms are seen when to reduce cardiovascular events in the acute setting of MI and
coronary endothelial damage occurs, and this can result in stroke, and in the secondary prevention of related ischemic
deleterious effects. Exposure of circulating blood to a thrombo- events (26). Risk reductions extend to secondary prevention in
genic surface results in platelet activation and aggregation as patients with unstable angina, coronary angioplasty, transient
well as release of vasoconstriction substances worsening under- ischemic attack (TIA), atrial fibrillation, and peripheral arterial
lying damage and ischemia. Thrombin converts fibrinogen to disease (26).
fibrin, and it further stimulates platelet activity. Once formed, a In addition to the desired therapeutic effects, inhibition of
thrombus may be degraded by endogenous or therapeutic COX-1 can erode protective gastric prostaglandin and cause
fibrinolysis or mechanically by opening the coronary artery direct gastric irritation, resulting in gastrointestinal bleeding.
via PCI; however, these processes may acutely promote further Bleeding events requiring hospitalization are rare (2/1000
platelet activation and thrombosis. patients) (27), and are more common with doses exceeding
Two types of agents discussed in this chapter act on 325 mg daily or in combination with other antiplatelet agents
different aspects of the thrombotic process. Antiplatelet medi- (28). Another worrisome adverse effect that appears to be dose
cations prevent platelet aggregation, while anticoagulants limit related is intracranial hemorrhage. Therefore, use of the lowest
further fibrin formation. These different sites of action allow for effective aspirin dose is vitally important. Maximal antiplatelet
additive, or potentially synergistic effects, resulting in greater effects of aspirin can be produced with as little as 30 mg once
efficacy with combination (antiplatelet and antithrombin) ther- daily (29). And in secondary prevention of TIA, it has been
apy. Combination therapy also increases bleeding risk. demonstrated that doses as low as 30 mg are as effective as
higher doses (282 mg daily) and have fewer adverse effects (30).
The disadvantage of low aspirin doses is that it may take up to
PHARMACOLOGIC FUNDAMENTALS two days to achieve maximal inhibition. Therefore, when
Antiplatelets immediate antiplatelet effect is desired, it is recommended
Antiplatelet medications have been a large focus of drug devel- that aspirin naı̈ve patients be administered 300 to 325 mg of
opment, and medications are now available to target various nonenteric coated aspirin (31). For chronic therapy, 81 mg daily
aspects of platelet function (Fig. 7.1). Antiplatelet medications is recommended.
can protect against platelet activation, aggregation, adhesion,
and platelet-induced vasoconstriction. Pharmacologic targets Thienopyridines
include inhibition of thromboxane A2 (TXA2) and adenosine Thienopyridines are another major pharmacologic oral antipla-
diphosphate (ADP) to decrease platelet activation and blockade telet class. Agents of this class irreversibly inhibit the binding of
of the glycoprotein (GP) IIb/IIIa receptor activity, the final ADP to the P2Y12 receptor on platelets, thereby preventing the
common pathway of platelet aggregation. transformation of a platelet to its activated form. Currently,
three such medications are approved by the U.S. Food and
Acetylsalicylic Acid Drug Administration (FDA): ticlopidine, clopidogrel, and pra-
Acetylsalicylic acid (aspirin) was the first antiplatelet therapy sugrel. The majority of thienopyridine-treated patients in the
used clinically. Aspirin exerts its antiplatelet effect via irrevers- United States are currently given clopidogrel because of its
ible acetylation of cyclooxygenase-1 (COX-1), which prevents favorable adverse effect profile compared with ticlopidine
arachidonic acid–induced production of TXA2. Thereby, both (32,33), the numerous clinical trials supporting its use in a
[2/7/010/7:23:1] [53–66]
Figure 7.1 Platelet activation is an important early step in the pathophysiology of atherothrombosis. Platelet activation involves (i) a shape
change in which the platelet membrane surface area is greatly increased; (ii) the secretion of proinflammatory, prothrombotic, adhesive, and
chemotactic mediators (release reaction), that propagate, amplify, and sustain the atherothrombotic process; and (iii) the activation of the GP
IIb/IIIa receptor from its inactive form. Multiple agonists including thromboxane A2 (TXA2), ADP, thrombin, serotonin, epinephrine, and
collagen, can activate the platelet and thus contribute toward establishing the environmental conditions necessary for atherothrombosis to
occur. Aspirin inhibits the production of thromboxane A2 by its effect on the enzyme cyclooxygenase (COX)-1. The ADP receptor antagonists
ticlopidine, clopidogrel, and prasugrel prevent the binding of ADP to its receptor. The effect of combining aspirin and clopidogrel is synergistic
in preventing platelet aggregation. Antithrombins such as unfractionated or low-molecular-weight heparin, hirudin, or bivalirudin are important
in interfering with both thrombin-induced platelet activation and coagulation. The GP IIb/IIIa receptor antagonists act at a later step in the
process by preventing fibrinogen mediated cross-linking of platelets, which have already become activated. Abbreviations: GP, glycoprotein;
ADP, adenosine diphosphate; ATP, adenosine triphosphate; PAI, plasminogen activator inhibitor; PDGF, platelet-derived growth factor; vWF,
von Willebrand factor. Source: Adapted from Ref. 24.
variety of settings, and the only recent FDA approval and noncardiac causes, stroke, severe hemorrhage, or peripheral
availability of prasugrel. vascular events was also significantly lower in the antiplatelet
Monotherapy with a thienopyridine is indicated in group, driven primarily by a reduction in hemorrhagic events.
patients with intolerance or contraindication to aspirin, or for Later, a clinical trial comparing aspirin alone, aspirin plus
the secondary prevention of ischemic stroke. However, the anticoagulation, and aspirin plus ticlopidine yielded similar
majority of thienopyridine use is in conjunction with aspirin results, with improved efficacy and less bleeding with dual
(dual antiplatelet therapy) following coronary artery stenting antiplatelet therapy compared with anticoagulation, also at
procedures or ACS. Short-term use of dual antiplatelet therapy 30 days (37). These trials established a minimum duration of
following placement of bare-metal stents was established in the dual antiplatelet therapy of 30 days following PCI with bare-
late 1990s when studies demonstrated that aspirin plus a metal stent placement. More recent observations have sug-
thienopyridine could decrease the incidence of life-threatening gested the value of longer durations of treatment (one year)
subacute stent thrombosis and major adverse cardiovascular with drug-eluting stents (DES).
events following PCI (34,35). Superior efficacy and safety of a Evidence to support dual antiplatelet therapy, beyond
dual antiplatelet therapy was seen when 257 patients undergo- when stent endothelization should be complete, comes from
ing PCI were randomized to either anticoagulation (heparin several landmark clinical trials. The Clopidogrel in Unstable
bridged to phenprocoumon with a target INR of 3.5–4.5 plus Angina to Prevent Recurrent Events (CURE) trial investigators
aspirin 100 mg twice daily) or antiplatelet therapy (ticlopidine examined the effects of clopidogrel (300 mg, followed by 75 mg
250 mg twice daily plus aspirin 100 mg twice daily) in the ISAR daily) in 12,562 patients with ACS without ST segment eleva-
trial (36). The primary efficacy end point at 30 days (a com- tions. The composite end point of cardiovascular death, MI or
posite of death, MI, bypass surgery, or repeat angioplasty) was stroke occurred less frequently in the clopidogrel arm compared
significantly reduced in the patients randomized to the anti- with placebo (9.3% and 11.4%, respectively) (34). Also, in the
platelet group (1.6% vs. 6.2%, RR 0.25, 95% CI 0.06–0.77). In CREDO trial subjects who underwent planned PCI and received
addition, the noncardiac composite end point of death from clopidogrel for one year had reductions in cardiovascular death,
[2/7/010/7:23:1] [53–66]
Table 7.6 Characteristics of Thienopyridine Antiplatelet Agents

Onset of Variability
% ADP antiplatelet in Hematologic Risk of
Drug Dose inhibition action Metabolism response toxicities Side effects bleeding
Ticlopidine MD: 250 mg 50–70% 4–7 days CYP450 3A þþþ þþ TTP, skin rash, þ
twice daily and 2C liver toxicity,
diarrhea
Clopidogrel LD: 300–600 mg 50–70% 1–8 hr CYP450 3A þþþ 0/þ Nonreversible þ
MD: 75 mg (following and 2C
once daily load)
Prasugrel LD: 60 mg 90% 30 min þþ 0/þ Nonreversible þþ
MD: 10 mg (following
once daily load)
MI, and strokes compared with placebo (35). Most recently, Interpatient variability in antiplatelet responses to thie-
13,608 patients with ACS and planned PCI were randomized to nopyridine therapy has been described and is of concern (46).
either prasugrel (60 mg, followed by 10 mg/day) or clopidogrel One theory that may partially explain the unpredictable
(300 mg, followed by 75 mg/day). Patients receiving prasugrel response is related to alterations in metabolism of prodrugs
had a significant reduction in the primary composite end point to their clinically effective metabolites. This occurs via metab-
of death from cardiovascular causes, nonfatal MI, or nonfatal olism by the liver cytochromes, primarily CYP3A4, CYP1A2
stroke throughout the follow-up period of over 14 months. and CYP2C19 isoenzymes (47). Several potentially significant
Increased efficacy of prasugrel over clopidogrel must be bal- drug interactions have been proposed with clopidogrel, all
anced with increased bleeding, including fatal bleeding, as seen involving this metabolic conversion in the liver. Although the
in this trial (38). These studies and evaluations following DES interaction between statins and clopidogrel has been shown to
(39–41), have led to the recommended use of aspirin plus a be clinically irrelevant, this was one of the first examined
thienopyridine for at least one year following ACS or PCI with (48–50). Atorvastatin was thought to compete with clopidogrel
stent placement. for CYP3A4 and pharmacodynamic data suggested a decreased
In addition to secondary prevention of atherosclerotic antiplatelet effect with coadministration (51). Recently,
vascular disease, clopidogrel has been investigated as peri- decreased antiplatelet effects have been observed with coad-
interventional therapy. On the basis of its ability to decrease ministration of proton-pump inhibitors (PPI) and clopidogrel,
platelet activation, use as an adjunct to, as well as an alternative likely due to inhibition of CYP2C19 (52). Studies of platelet
to more traditional options, such as GP IIb/IIIa inhibitors has assays and observational data from large clinical trials resulted
been proposed. A substudy of the CURE trial analyzed the in the FDA issuing a warning of decrease antiplatelet effect
2658 subjects who underwent PCI (42). All patients received when using the combination of PPIs and clopidogrel. In con-
open-label aspirin and thienopyridine for four weeks after PCI. trast however, the preliminary results of the only randomized
Importantly, clopidogrel or placebo was started for a median of placebo-controlled clinical trial show no worsening in outcome
six days prior to PCI. Overall, pretreatment with clopidogrel was among patients receiving omeprazole and clopidogrel. In addi-
associated with a reduction in the composite end point of car- tion, smoking, a known inducer of the CYP1A2 isoenzyme, may
diovascular death, MI, or urgent target vessel revascularization enhance the antiplatelet effects of clopidogrel by enzyme induc-
(TVR). Similarly, a substudy of the CREDO trial demonstrated tion (53). The elucidation of clinical implications of these
that patients who received 300 mg loading dose greater than interactions warrants further study. Nonetheless, it appears
15 hours prior to PCI had a decrease in the 28-day composite end that fewer drug interactions and less interpatient variability is
point of death, MI, or urgent TVR (43). Table 7.6 outlines impor- seen with use of prasugrel, resulting in faster, greater, and more
tant characteristics of the individual thienopyridines. consistent inhibition of ADP-induced platelet aggregation than
Further evidence supporting clopidogrel pretreatment with clopidogrel (54).
(particularly with a high loading dose) as an alternative to
GP IIb/IIIa in low-risk patients comes from the ISAR-REACT 1 GP IIb/IIIa Receptor Antagonists
(44) and ISAR-REACT 2 trials (45). The former evaluated the GP IIb/IIIa receptor antagonists block the final common pathway of
use of abciximab versus placebo in 2159 undergoing elective platelet aggregation by preventing fibrinogen binding and cross-
PCI, all of whom had received 600 mg of clopidogrel in linking of platelets at the aIIab3-receptor. Interference with this
advance. No difference was seen in the composite end point process results in inhibition of approximately 80% of platelet
of death, MI, or urgent TVR within 30 days after randomization aggregation function (55). Three such medications are FDA
(RR 1.05, 95% CI 0.69–1.59, p ¼ 0.82). ISAR-REACT 2 had a approved for use in conjunction with aspirin and antithrombotic
similar trial design applied to 2022 patients with non–ST ele- therapy, all are administered intravenously and for a limited
vation ACS undergoing PCI. In this higher-risk patient popu- period of time, providing potent antiplatelet therapy for the most
lation, abciximab reduced the composite end point of death, critical interval following ACS diagnosis and/or stent place-
MI, or urgent TVR at 30 days when compared with placebo (RR ment. As previously mentioned, the use of clopidogrel pretreat-
0.75, 95% CI 0.58–0.97, p ¼ 0.03). These studies demonstrate that ment and newer antithrombotic regimens has lessened the role
clopidogrel pretreatment is sufficient for low-risk PCI, how- of GP IIb/IIIa inhibition, however, they remain important for
ever, in patients in whom adequate pretreatment is not possi- intermediate- and high-risk ACS and higher-risk PCI cases.
ble, or among high-risk patients (i.e., troponin positive), GP IIb/IIIa inhibitors have been widely tested. The
alternate antiplatelet strategies may be needed. majority of benefit is seen among ACS patients who undergo
[2/7/010/7:23:1] [53–66]
early PCI, those with dynamic ST segment changes, and tor, and eptifibatide is a synthetic cyclic heptapeptide. Both are
individuals with elevated troponin levels (56). Benefit has inherently less likely to cause hypersensitivity reactions than
also been established among patients with high-risk features abciximab (61). Additionally, these agents are preferred in
such as diabetes or home aspirin therapy (57). Additionally, use patients at increased risk of bleeding, because of reversible
of these agents prior to planned PCI has been clearly estab- antiplatelet activity, shorter duration of action, and lower inci-
lished (57–59). However, much of the trial data came from dence of severe thrombocytopenia (66). These agents have been
studies prior to the routine inclusion of thienopyridines and in studied with favorable results in PCI-patient populations sim-
lower risk patient populations no benefit has been observed. ilar to those in trials of abciximab (67–69), but also have a role in
Mixed outcomes have resulted in the practice of preferentially ACS patients managed without PCI.
treating only the highest risk patients with upstream GP IIb/ Several trials have evaluated the adjunctive use of small-
IIIa antagonists, although no study has validated this approach. molecule GP IIb/IIIa inhibitors as medical therapy for ACS in
Individual differences in the GP IIb/IIIa antagonists and sup- patients receiving aspirin and heparin. Among the earliest of
portive clinical trial data can be used to guide which therapy to these was the PURSUIT trial, where treatment with eptifibatide
use depending on the specific patient and clinical scenario. was continued for up to 72 to 96 hours (59). The eptifibatide
Several important pharmacologic factors define the use of treated patients experienced a 1.5% absolute risk reduction in
abciximab. In contrast to other GP IIb/IIIa inhibitors, abciximab the 30-day composite end point of death or nonfatal MI. The
is a monoclonal antibody blocking GP IIb/IIIa receptors. Fol- effect was pronounced in patients undergoing PCI during drug
lowing intravenous bolus, antiplatelet effects occur within administration. Trials evaluating the use of tirofiban in a similar
minutes and return to normal within 48 hours in most cases non–ST elevation ACS patient population showed similar ben-
(60). The general irreversibility and slower offset compared efit, reducing early cardiovascular ischemic events (70,71). How-
with small-molecule GP IIb/IIIa inhibitors limits abciximab ever, subjects in these trials had limited “upstream” exposure
when there is a high likelihood of bleeding complications or with median time to PCI of <6 hours, and conflicting results
surgical intervention. Since abciximab is an immune fragment, were seen in the recently published EARLY-ACS trial (72).
immune mediated thrombocytopenia and other responses, Evaluations of upstream GP IIb/IIIa inhibitor use in
while overall rare, are also more common (61). patients experiencing STEMI and undergoing primary PCI
Abciximab is only approved for use in patients undergo- have yielded mixed results. Both the ADMIRAL and RELAx-
ing or planned to undergo PCI. In the first evaluation of GP AMI trials demonstrated that early abciximab administration
IIb/IIIa inhibitors for PCI, abciximab (0.25 mg/kg bolus plus resulted in improved coronary angiographic findings at the
10 mcg/min infusion for 12 hours) resulted in a significant time of PCI (73,74). However, clinical outcomes were not
reduction in the primary composite end point (death, MI, improved by the early addition of abciximab in the FINESSE
revascularization, stent placement, or IABP insertion). This trial (75). In the placebo-controlled second Ongoing Tirofiban in
study enrolled 2099 patients with unstable angina, high-risk Myocardial Evaluation (ON-TIME 2) trial, the prehospital
features, or undergoing angioplasty or atherectomy for MI (62). administration of high-bolus-dose tirofiban significantly
Benefit was subsequently observed in a lower risk, elective PCI improved ST segment resolution both before and after primary
cohort. This evaluation of 2792 patients was stopped early PCI in patients with acute STEMI (76).
when abciximab resulted in a 6.5% absolute risk reduction in Overall, meta-analysis including multiple trials, enrolling
the primary efficacy end point (death, MI, or urgent TVR) (63). 31,402 patients with ACS without planned PCI showed a
Later, once implementation of PCI with stent placement modest benefit of GP IIb/IIIa inhibitors compared with controls
became widespread, the EPISTENT trial sought to investigate (77). Considering a 30-day end point, Boersma and colleagues
abciximab’s role in stenting procedures (64). reported a 9% relative risk reduction (10.8% vs. 11.8%,
EPISTENT enrolled 2399 patients scheduled to undergo p ¼ 0.015) in death or myocardial infarction. The authors
elective or urgent PCI with stent placement to one of three concluded that upstream IIb/IIIa inhibitors might be started
arms: stenting plus placebo, stenting plus abciximab, or balloon and continued until an invasive approach was made. With
angioplasty plus abciximab. All patients received heparin and mixed trial results, growing emphasis placed on improving the
ASA. Again, theinopyridines were routinely administered time from symptom onset to reperfusion, rapid-acting thieno-
before PCI with a loading dose. Both abciximab arms resulted pyridine strategies, and alternative anticoagulation regimens
in a statistically significant reductions in the composite of death such as bivalirudin, the upstream use of GP IIb/IIIa inhibitors
from any cause, MI, or urgent revascularization relative to have most recently lost some appeal.
placebo plus stenting (64). Abciximab was separately observed Contraindications to GP IIb/IIIa therapy are similar
to be effective during primary PCI for treatment of ST segment among abciximab, eptifibatide, and tirofiban. Given their abil-
elevation MI. A meta-analysis of three trials (ISAR-2, ADMI- ity to cause profound thrombocytopenia (abciximab > eptifiba-
RAL, and ACE) compared 550 patients on abciximab with 551 tide > tirofiban), history of such reactions should result in a
patients receiving placebo. Abciximab use resulted in a 37% clear contraindication to readministration. Also, patients with
reduction in the composite of death or reinfarction (p ¼ 0.008), baseline platelet counts <100,000 should not receive GP IIb/IIIa
with a nonsignificant increase in bleeding (65). More recent therapy. Recent history of ischemic stroke (within three
studies with high loading doses of clopidogrel have not months) and any history of hemorrhagic stroke warrant careful
observed a reduction in ischemic events, though these obser- evaluation of risk verses benefits of use. Also, with bleeding a
vations have not been definitive. limiting side effect, administration in patients with trauma,
Eptifibatide and tirofiban are both small-molecule agents. recent surgery, and bleeding disorders is questionable. Finally,
Although different from abciximab in pharmacology, the end careful attention should be paid to the dosing of these medi-
result of preventing fibrinogen and Von Willebrand factor cations, particularly given the weight-based dosing and renal
binding to the GP IIb/IIIa receptor is similar. Tirofiban is a adjustments recommended with several of the medications.
highly specific nonpeptide peptidomimetic GP IIb/IIIa inhibi- Table 7.7 shows the characteristics of the GP IIb/IIIa inhibitors.
[2/7/010/7:23:1] [53–66]
Table 7.7 Characteristics of the Glycoprotein IIb/IIIa Inhibitors

Abciximab Eptifibatide Tirofiban Lamifiban
Brand name ReoPro Integrilin Aggrastat –
Structure Antibody Fab fragment Cyclic heptapeptide Nonpeptide Nonpeptide
Molecular weight 47.6 0.832 0.495 0.468
(kD)
Plasma half-life 10–30 min *2.5 hr *2 hr *2 hr
Excretion Unknown *50% renal 39–69% renal 90% renal
Approved PCI ACS and PCI (unstable angina ACS and PCI (unstable Not approved
indications Refractory unstable angina and non–Q wave MI) angina and non–Q
when PCI is planned within wave MI)
24 hr
Recommended For PCI: For ACS: For ACS: Not approved
dose 0.25 mg/kg bolus, 180 mcg/kg bolus, 2.0 mcg/kg/min 0.4 mcg/kg/min
0.125 mcg/kg/min infusion 72–96 hr (PURSUIT dose) 30 min, then
(maximum 10 mcg/min) For PCI: 0.1 mcg/kg/min
infusion 12 hr 180 mcg/kg bolus 2, 48–108 hr
2.0 mcg/kg/min infusiona 18 hr
a
Eptifibatide infusion should be decreased to 1 mcg/kg/min if creatinine clearance <50 ml/min.
Abbreviations: PCI, percutaneous coronary intervention; ACS, acute coronary syndrome.
Newer Agents (Ticagrelor, Cangrelor, and SCH 530348) addition to the thrombogenicity of the acute circumstance, intro-
Newer agents (ticagrelor, cangrelor, and SCH 530348) are in var- duction of a foreign body (e.g., coronary guidewires) and
ious stages of development because of the desire to improve mechanical injury caused by the procedure can easily result in
cardiovascular outcomes and lessen antiplatelet-related com- thrombus formation or clot progression. Use of anticoagulation
plications. A limiting factor to the use of thienopyridines is can reduce clot progression, acute stent thrombosis, and subse-
their irreversible nature. Ticagrelor is an oral ADP receptor quent ischemic events. Unfractionated heparin (UFH), low–
P2Y12 antagonist that has been evaluated in several clinical molecular-weight heparin (LMWH), and bivalirudin are the
trials. In dose ranging studies of ticagrelor, 90 or 180 mg twice most commonly used anticoagulants in PCI. Agent selection is
daily compared with standard doses of clopidogrel has resulted based on clinical data, diagnosis, and patient characteristics.
in dose-dependent and greater inhibition of platelet aggrega- However, benefits must be balanced with increases in bleeding.
tion (79). In the recently published PLATO trial, ticagrelor was Table 7.8 enumerates currently available antithrombotic agents
studied in patients with ACS with and without ST segment for PCI.
elevation. Ticagrelor reduced the primary composite end point
(CV death, MI or stroke) by 1.9% compared with clopidogrel, Heparin
without an increase in the rate of major bleeding (80). While Heparin is a heterogenous mucopolysaccharide with complex
this P2Y12 inhibitor is showing great potential, another ADP effects on the coagulation pathway (83). Heparin is a naturally-
inhibitor has experienced somewhat different results. Cangre- occurring anticoagulant produced by human basophils and
lor’s novel features include intravenous formulation, short half- mast cells (83); however, most of the pharmaceutically available
life (3–5 minutes), and reversibility. Two large phase III trials to heparin in the United States is derived from porcine mucosal
evaluate cangrelor where recently halted, when efficacy end tissue. The primary pharmacologic effect is thought to be
points appeared that the drug would not show enough benefit driven by the binding of heparin to antithrombin and simulta-
to obtain FDA approval. A third antiplatelet therapy with a neous binding of heparin to thrombin. This promotes the effects
novel mechanism is currently under development. SCH 530348 of antithrombin III, which prevents the binding of fibrinogen to
is an oral platelet protease-activated receptor 1 (PAR-1) antag- thrombin, and hence is an anticoagulant. Importantly, it also
onist, and unlike other “antiplatelet” therapies available this exerts indirect antiplatelet effects by binding to and inhibiting
medication prevents thrombin-induced platelet activation. Von Willebrand factor. Additionally, UFH binds to plasma
SCH 530348 is being evaluated for the prevention of athero- proteins, endothelial cells, and macrophages, which in turn
thrombotic events in patients with ACS, peripheral arterial inactivates heparin. These wide-ranging activities result in
disease, or a history of stroke. The tolerability and safety of variable anticoagulant effects, and as such, therapy with UFH
SCH 530348 was tested in patients undergoing coronary angiog- requires close monitoring and dosing adjustments during use.
raphy with possible PCI. Study investigators found that the Heparin has been the gold standard for anticoagulants in
medication was well tolerated with no increase in TIMI bleed- the catheterization laboratory; however, no prospective trials
ing, even when added to aspirin and clopidogrel therapy (81). have clearly defined the optimal level of anticoagulation nec-
Two large-scale clinical trials evaluating SCH 530348, one for essary during PCI, and concomitant antithrombotic regimens
the treatment of ACS (82), and one for secondary prevention, are continue to evolve. Monitoring and titration of UFH during PCI
underway. can be facilitated through point-of-care testing of the activated
clotting times (ACT) (84). Historically, the standard heparin
Anticoagulants regimen has been 100 units/kg bolus with additional weight-
Anticoagulation during PCI is of vital importance. The previously based boluses to achieve and maintain an ACT of 250 to
mentioned antiplatelet therapies are all given on a background 350 seconds, depending on concomitant therapies (85).
of anticoagulation during PCI and often peri-procedurally. In Although lacking prospective validation, retrospective data
[2/7/010/7:23:1] [53–66]
Table 7.8 Antithrombotic Therapy for PCI

Parameter UFH Enoxaparin Fondaparinux Bivalirudin Argatroban
Route IV IV, subcutaneous IV, subcutaneous IV IV
Clotting factors IIa, IX, Xa IIa, Xa Xa DTI DTI
inhibited
Metabolism Unknown Hepatic (desulfation) NA Hepatic and Hepatic
blood
Elimination Renal Renal Renal (20% Renal (16%
(10% unchanged) (77% unchanged) unchanged) unchanged)
Half-life 1 hr 2–4 hr 17–21 hr 25 min 30–51 min
Dose 60–100 units/kg 0.3–1.0 mg/kg IVa 0.5–1.0 mg/kg IVa 0.75 mg/kg IV then 350 mcg/kg IV then
IV prn 1.75 mg/kg/hr 25 mcg/kg/min
Coagulation ACT (200 with None required, None required, ACT ACT
monitoring IIb/IIIa and plasma antifactor plasma antifactor (>225) (300–450)
and target for 300–350 Xa Xa
PCI alone) (if needed) (if needed)
Reversibility Protamine 66% by protamine, FFP FFP FFP
FFP
a
dependent on the timing of the last subcutaneous dose and concomitant IIb/IIIa use.
Abbreviations: PCI, percutaneous coronary intervention; ACT, activated clotting times; DTI, direct thrombin (IIa) inhibitors.
show that this approach is reasonable. One retrospective study UFH. Several LMWH are available though the majority of
showed short-term ischemic events were more likely (6.6% evaluations in PCI has involved enoxaparin. Like UFH, enox-
compared with 11.1%) in patients with ACTs of 171 to 195 aparin may be used to treat ACS with or without planned PCI
compared with those with higher ACT ranges of 350 to 375 and to prevent recurrent vessel occlusion following thrombo-
seconds (86). This is balanced with the knowledge that maxi- lytic reperfusion.
mum ACT is predictive of bleeding complications (87). Impor- Several clinical trials have compared UFH and enoxa-
tantly, sheath removal is safest once ACTs fall below 150 to 180. parin in the medical treatment of ACS. As examples, the two
When UFH is administered with GP IIb/IIIa inhibitors, drugs were compared in patients with NSTE ACS in both the
lower ACT values of *200 seconds result in similar reductions ESSENCE trial and the A to Z trial. Both trials were open-
of ischemic events with less risk bleeding risk (85,88). A differ- labeled, randomized, noninferiority designs comparing stan-
ent approach is used when UFH is started upstream, and used dard dosing regimens of UFH with enoxaparin 1 mg/kg
for the medical management of ACS, or when used for pre- subcutaneously every 12 hours. In ESSENCE enoxaparin was
vention of reocclusion following fibrinolysis. In these scenarios, found to be superior to UFH by reducing the combined end
UFH is traditionally administered as a maximal 5000 unit bolus point of death, MI, or recurrent angina (94). Later, in the
followed by 1000 units/hr for up to 48 hours (89). Rather than tirofiban arm of the A to Z trial, enoxaparin was shown to be
ACT, the activated partial thromboplastin time (aPTT) is noninferior to UFH when given in addition to aspirin and
utilized for monitoring with a goal of 50 to 70 seconds or 1.5 tirofiban (95). On the basis of these trial results, either UFH
to 2 times baseline (90). Whether UFH is started before proce- or enoxaparin are viable options for medical management,
dure or only utilized during PCI, immediate discontinuation though guideline preference is given to enoxaparin.
following successful PCI is recommended. Continuation has LMWH has also been compared with UFH in the setting
shown no benefit and results in higher rates of access site of PCI. The Superior Yield of the New Strategy of Enoxaparin,
bleeding (91,92). Given the many idiosyncrasies with UFH, it Revascularization and GP IIb/IIIa Inhibitors (SYNERGY) trial
is far from ideal as an anticoagulant. randomized 9978 higher-risk ACS patients with planned early
invasive treatment to either UFH or enoxaparin. The results
Low–Molecular-Weight Heparins (LMWH) showed noninferiority, with no difference in the primary end
LMWH are heparin salts and are about one-third of the molec- point (death or MI), subacute stent thrombosis, or unsuccessful
ular weight of UFH. Mechanistically, these shorter-chain poly- procedures (96). These finding are in agreement with results of
saccharides bind to antithrombin and thrombin, but the the subgroup analysis of patients who received PCI in the A to
binding to thrombin is to a lesser extent than UFH. Rather, Z trial (95). However, the SYNERGY trial did yield another
the majority of anticoagulant effect comes from binding to interesting and important finding. Bleeding was increased in
factor Xa, the catalyst for conversion of prothrombin to throm- patients who received enoxaparin before PCI and crossed over
bin. As a result, LMWH has only a minimal effect on aPTT and to receive UFH during PCI. Therefore, “crossover” should be
ACT, and monitoring requires measurement of anti-Xa activity. avoided and in patients who present to a catheterization pro-
Bedside monitoring of anti-Xa levels is not routine; therefore, in cedure already receiving enoxaparin, subsequent doses should
patients receiving LMWH during PCI, dosing recommenda- be based on the timing of the previous dose. If the last dose of
tions should be based on available evidence from clinical trials enoxaparin was <8 hours prior to PCI, no additional anti-
(93). LMWH have several important pharmacologic differences coagulant is needed, if received 8 to 12 hours prior to PCI,
compared with UFH, including an increased bioavailability, a 0.3 mg/kg IV enoxaparin should be given, and finally, if
prolonged route of elimination, and a more consistent anti- enoxaparin was administered >12 hours prior, a full-dose is
coagulant effect. Therefore, following IV or SQ administration, needed or conventional therapy is indicated (97).
they offer a reasonably predictable level of anticoagulation and The safety of enoxaparin in elective PCI was evaluated
longer lasting antithrombotic effect than that achieved with in the STEEPLE trial (98). Patients undergoing PCI were
[2/7/010/7:23:1] [53–66]
randomized to 0.5 mg/kg IV, 0.75 mg/kg IV, or UFH adjusted bivalirudin (0.75 mg/kg bolus plus 1.75 mg/kg/hr for the
for ACT. The primary end point, incidence of major or minor duration of PCI) or UFH with a GP IIb/IIa inhibition (105).
bleeding not related to CABG, was decreased in the 0.5 mg/kg The authors concluded that bivalirudin was noninferior to
arm. There was no difference in bleeding between the patients heparin and GP IIb/IIIa inhibition with regard to suppression
receiving UFH and the higher-dose of enoxaparin. This trial of the primary efficacy end points (death, MI, or TVR) and was
supports the use of a single intravenous bolus of 0.5 mg/kg associated with less bleeding (2.4% vs. 4.1%; p < .001) (105).
enoxaparin to subjects undergoing elective PCI; however it Provisional GP IIb/IIIa inhibitor therapy was administered to
was not particularly large enough to assess efficacy of these 7.2% of patients in the bivalirudin arm. This trial extended
modalities. bivalirudin use to patients undergoing elective PCI. Bivalirudin
would later demonstrate efficacy when initiated upstream from
Fondaparinux PCI, in the management of NSTEMI, unstable angina, and
Fondaparinux is a synthetic pentasaccharide that is an antith- STEMI (106–108).
rombin-dependent indirect inhibitor of activated factor X (Xa). In the Acute Catheterization and Urgent Intervention
Its use as an alternative to heparin, in both the medical man- Triage Strategy (ACUITY) trial, bivalirudin (with and without
agement of ACS and antithrombotic protection post fibrinolysis GP IIb/IIIa inhibition) was compared with heparin plus a GP
has been evaluated. In the Fifth Organization to Assess Strat- IIb/IIIa inhibitor (106). The trial enrolled 13,819 patients with
egies in Ischemic Syndromes (OASIS)-5 trial, 20,078 patients moderate to high-risk acute coronary syndromes and planned
experiencing ACS were randomized to enoxaparin or fonda- early interventions. Stone et al. reported that a strategy of
parinux. The primary short-term efficacy end point of death, bivalirudin alone reduced rates of major bleeding (3.0% vs.
MI, or stroke occurred at a similar frequency in both groups; 5.7%, p < 0.001) with similar efficacy (7.8% and 7.3%, p ¼ 0.32)
with fondaparinux recipients experiencing 50% fewer bleeding to heparin plus GP IIb/IIIa inhibition. Later, the HORIZONS-
events and 17% fewer mortalities at 30 days. In 2007, Mehta AMI trial addressed the use of bivalirudin in patients with ST
et al. published results from the planned analysis of the 6238 segment elevation MI undergoing primary PCI (107). The end
patients who underwent PCI (99). Short-term rates of ischemic points studied were major bleeding and the combination of
events were similar, and major bleeding was reduced by one- death, reinfarction, TVR, and stroke. Anticoagulation with
half; however, patients in the fondaparinux group experienced bivalirudin alone was compared with heparin plus GP IIb/
higher rates of catheter thrombosis (0.9% compared with 0.4%). IIIa inhibitors and resulted in a significant decrease in bleeding
Because of the risk of this potentially devastating complication, and similar efficacy with regard to ischemic end points.
fondaparinux should not be used as the sole anticoagulant to Considering the findings of ACUITY and HORIZONS-
support PCI (85). Furthermore, fondaparinux should be AMI, bivalirudin is an attractive approach for patients under-
avoided in patients when an early invasive strategy is planned. going elective or urgent PCI, particularly when adequate anti-
Several new oral anticoagulants are being developed, with the platelet therapy with a thienopyridine is present. In these
oral direct factor Xa inhibitor apixaban being evaluated in the settings, with over 10,000 patient experiences in clinical trials,
treatment of patients with recent ST elevation or non–ST ele- bivalirudin therapy continues to be noninferior to heparin plus
vation acute coronary syndrome. The phase II trial, APPRAISE- GP IIb/IIIa inhibitor therapy with substantial reductions in
1, showed dose-related increases in bleeding and a decrease in bleeding.
ischemic events (100). Apixiban 5 mg twice daily is being
evaluated in a phase III study in a similar patient population.
CONCLUSION
Direct Thrombin Inhibitors Given there are many antiplatelet, antithrombin, and antifibrin
Direct thrombin inhibitors (DTI) currently available in the United drugs available for PCI, polypharmacy-based procedures are
States include bivalirudin, argatroban, and lepirudin. DTI standard of care with literally hundreds of potential drug
inhibit soluble and clot-bound thrombin without involvement combinations. The overall pharmacologic management of
of antithrombin. Another potential advantage over heparin is patients during PCI can be remarkably complex. Hemody-
that DTI do not promote platelet activity, rather the degree of namic management is undertaken with the ultimate goal of
thrombin inhibition at high doses, decreases platelet activation. patient stabilization to allow potentially life-saving procedures
The earliest experience with DTI in PCI was mostly with to take place and ultimately to improve outcomes and survival.
argatroban, however the majority of prospective randomized It is also vitally important to prevent complications using
evaluations has utilized bivalirudin. In 1999, patients with appropriate periprocedural pharmacotherapy. Development
known or suspected HIT and requiring PCI were given arga- of the ideal antithrombotic and antiplatelet regimen continues
troban rather than heparin. In this small, single-armed study, to be a goal for the cardiology community. New therapies are in
use of the DTI argatroban resulted in adequate anticoagulation, development and may help increase efficacy and limit adverse
and bleeding was minimal (101). Later, when GP IIb/IIIa ther- effects in this at-risk patient population.
apy was more common, a retrospective evaluation of argatro-
ban yielded similar results (102). Now clinical trial data
(103,104) and extensive experience with bivalirudin in PCI REFERENCES
1. Shoemaker WC, Appel PL, Kram HB, et al. Prospective trial of
support its use in high-risk patients with HIT requiring PCI.
supranormal values of survivors as therapeutic goals in high-risk
Favorable experiences seen in these patients, and inherent
surgical patients. Chest 1988; 94(6):1176–1186.
weaknesses of heparin therapy resulted in evaluation of biva- 2. Tuchschmidt J, Fried J, Astiz M, et al. Elevation of cardiac output
lirudin in patients without contraindication to heparinoids. and oxygen delivery improves outcome in septic shock. Chest
The Randomized Evaluation in PCI Linking Angiomax 1992; 102(1):216–220.
to Reduced Clinical Events (REPLACE-2) trial randomized 3. Effects of enalapril on mortality in severe congestive heart fail-
6010 patients undergoing urgent or elective PCI to intravenous ure. Results of the Cooperative North Scandinavian Enalapril
[2/7/010/7:23:1] [53–66]
Survival Study (CONSENSUS). The CONSENSUS Trial Study 24. Mehta SR, Yusuf S. Short- and long-term oral antiplatelet therapy
Group. N Engl J Med 1987; 316(23):1429–1435. in acute coronary syndromes and percutaneous coronary inter-
4. Mullens W, Abrahams Z, Francis GS, et al. Sodium nitroprusside vention. J Am Coll Cardiol 2003; 21:79S–88S.
for advanced low-output heart failure. J Am Coll Cardiol 2008; 25. Patrono C, Ciabattoni G, Patrignani P, et al. Clinical pharmacol-
52(3): 200–207. ogy of platelet cyclooxygenase inhibition. Circulation 1985; 72(6):
5. Opie L, Gersch B, eds. Drugs for the Heart. 6th ed. Philadelphia, 1177–1184.
PA: Elsevier, 2005: 161–162. 26. Antithrombotic Trialists’ Collaboration. Collaborative meta-
6. Silver MA, Horton DP, Ghali JK, et al. Effect of nesiritide versus analysis of randomised trials of antiplatelet therapy for prevention
dobutamine on short-term outcomes in the treatment of patients of death, myocardial infarction, and stroke in high-risk patients.
with acutely decompensated heart failure. J Am Coll Cardiol BMJ 2002; 324(7329):71–86.
2002; 39(5):798–803. 27. Farrell B, Godwin J, Richards S, et al. The United Kingdom
7. Cuffe MS, Califf RM, Adams KF Jr, et al. Short-term intravenous transient ischaemic attack (UK-TIA) aspirin trial: final results.
milrinone for acute exacerbation of chronic heart failure: a J Neurol Neurosurg Psychiatry 1991; 54(12):1044–1054.
randomized controlled trial. JAMA 2002; 287(12):1541–1547. 28. Roderick PJ, Wilkes HC, Meade TW. The gastrointestinal toxicity
8. Unverferth DV, Magorien RD, Lewis RP, et al. Long-term benefit of aspirin: an overview of randomised controlled trials. Br J Clin
of dobutamine in patients with congestive cardiomyopathy. Am Pharmacol 1993; 35(3):219–226.
Heart J 1980; 100(5):622–630. 29. Patrignani P, Filabozzi P, Patrono C. Selective cumulative inhi-
9. Marino P, ed. The ICU Book. 3rd ed. Philadelphia, PA: Lippincott bition of platelet thromboxane production by low-dose aspirin in
Williams & Wilkins, 2007:176, 264–267. healthy subjects. J Clin Invest 1982; 69(6):1366–1372.
10. Kern M, ed. SCAI Interventional Cardiology Board Review 30. A comparison of two doses of aspirin (30 mg vs. 283 mg a day) in
Book. 1st ed. Philadelphia, PA: Lippincott Williams & Wilkins, patients after a transient ischemic attack or minor ischemic
2006. stroke. The Dutch TIA Trial Study Group. N Engl J Med 1991;
11. GISSI-3: effects of lisinopril and transdermal glyceryl trinitrate 325(18):1261–1266.
singly and together on 6-week mortality and ventricular function 31. Smith SC Jr, Feldman TE, Hirshfeld JW Jr, et al. ACC/AHA/
after acute myocardial infarction. Gruppo Italiano per lo Studio SCAI 2005 guideline update for percutaneous coronary interven-
della Sopravvivenza nell’infarto Miocardico. Lancet 1994; 343(8906): tion-summary article: a report of the American College of Car-
1115–1122. diology/American Heart Association Task Force on practice
12. Flather M, Pipilis A, Collins R, et al. Randomized controlled trial guidelines (ACC/AHA/SCAI writing committee to update the
of oral captopril, of oral isosorbide mononitrate and of intrave- 2001 guidelines for percutaneous coronary intervention). J Am
nous magnesium sulphate started early in acute myocardial Coll Cardiol 2006; 47(1):216–235.
infarction: safety and haemodynamic effects. ISIS-4 (Fourth Inter- 32. Bhatt DL, Bertrand ME, Berger PB, et al. Meta-analysis of
national Study of Infarct Survival) Pilot Study Investigators. Eur randomized and registry comparisons of ticlopidine with clopi-
Heart J 1994; 15(5):608–619. dogrel after stenting. J Am Coll Cardiol 2002; 39(1):9–14.
13. Dell’Italia LJ, Starling MR, Blumhardt R, et al. Comparative 33. Bertrand ME, Rupprecht HJ, Urban P, et al. Double-blind study
effects of volume loading, dobutamine, and nitroprusside in of the safety of clopidogrel with and without a loading dose in
patients with predominant right ventricular infarction. Circula- combination with aspirin compared with ticlopidine in combi-
tion 1985; 72(6):1327–1335. nation with aspirin after coronary stenting: the Clopidogrel
14. Berne RM. The role of adenosine in the regulation of coronary Aspirin Stent International Cooperative Study (CLASSICS). Cir-
blood flow. Circ Res 1980; 47(6):807–813. culation 2000; 102(6):624–629.
15. Christensen CW, Rosen LB, Gal RA, et al. Coronary vasodilator 34. Yusuf S, Zhao F, Mehta SR, et al. Effects of clopidogrel in addition
reserve. Comparison of the effects of papaverine and adenosine to aspirin in patients with acute coronary syndromes without ST-
on coronary flow, ventricular function, and myocardial metabo- segment elevation. N Engl J Med 2001; 345(7):494–502.
lism. Circulation 1991; 83(1):294–303. 35. Steinhubl SR, Berger PB, Mann JT III, et al. Early and sustained
16. Leier CV, Heban PT, Huss P, et al. Comparative systemic and dual oral antiplatelet therapy following percutaneous coronary
regional hemodynamic effects of dopamine and dobutamine in intervention: a randomized controlled trial. JAMA 2002; 288(19):
patients with cardiomyopathic heart failure. Circulation 1978; 2411–2420.
58(3 pt 1):466–475. 36. Schomig A, Neumann FJ, Kastrati A, et al. A randomized compar-
17. Viskin S. Long QT syndromes and torsade de pointes. Lancet ison of antiplatelet and anticoagulant therapy after the placement
1999; 354(9190):1625–1633. of coronary-artery stents. N Engl J Med 1996; 334(17):1084–1089.
18. Bellomo R, Chapman M, Finfer S, et al. Low-dose dopamine 37. Leon MB, Baim DS, Popma JJ, et al. A clinical trial comparing
in patients with early renal dysfunction: a placebo-controlled three antithrombotic-drug regimens after coronary-artery stent-
randomised trial. Australian and New Zealand Intensive Care ing. Stent Anticoagulation Restenosis Study Investigators. N Engl
Society (ANZICS) Clinical Trials Group. Lancet 2000; 356(9248): J Med 1998; 339(23):1665–1671.
2139–2143. 38. Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versus
19. Marik PE, Iglesias J. Low-dose dopamine does not prevent acute clopidogrel in patients with acute coronary syndromes. N Engl J
renal failure in patients with septic shock and oliguria. NORA- Med 2007; 357(20):2001–2015.
SEPT II Study Investigators. Am J Med 1999; 107(4):387–390. 39. Lagerqvist B, James SK, Stenestrand U, et al. Long-term out-
20. Kellum JA, M Decker J. Use of dopamine in acute renal failure: a comes with drug-eluting stents versus bare-metal stents in
meta-analysis. Crit Care Med 2001; 29(8):1526–1531. Sweden. N Engl J Med 2007; 356(10):1009–1019.
21. Bassi G, Radermacher P, Calzia E. Catecholamines and vaso- 40. Stone GW, Moses JW, Ellis SG, et al. Safety and efficacy of
pressin during critical illness. Endocrinol Metab Clin North Am sirolimus- and paclitaxel-eluting coronary stents. N Engl J Med
2006; 35(4):839–857, x. 2007; 356(10):998–1008.
22. Lindner KH, Dirks B, Strohmenger HU, et al. Randomised 41. Win HK, Caldera AE, Maresh K, et al. Clinical outcomes and
comparison of epinephrine and vasopressin in patients with stent thrombosis following off-label use of drug-eluting stents.
out-of-hospital ventricular fibrillation. Lancet 1997; 349(9051): JAMA 2007; 297(18):2001–2009.
535–537. 42. Mehta SR, Yusuf S, Peters RJ, et al. Effects of pretreatment with
23. Barrett LK, Singer M, Clapp LH. Vasopressin: mechanisms of clopidogrel and aspirin followed by long-term therapy in
action on the vasculature in health and in septic shock. Crit Care patients undergoing percutaneous coronary intervention: the
Med 2007; 35(1):33–40. PCI-CURE study. Lancet 2001; 358(9281):527–533.
[2/7/010/7:23:1] [53–66]
43. Steinhubl SR, Berger PB, Brennan DM, et al. Optimal timing for 60. Mascelli MA, Lance ET, Damaraju L, et al. Pharmacodynamic
the initiation of pre-treatment with 300 mg clopidogrel before profile of short-term abciximab treatment demonstrates
percutaneous coronary intervention. J Am Coll Cardiol 2006; prolonged platelet inhibition with gradual recovery from GP
47(5):939–943. IIb/IIIa receptor blockade. Circulation 1998; 97(17):1680–1688.
44. Kastrati A, Mehilli J, Schuhlen H, et al. A clinical trial of abciximab 61. Topol EJ, Moliterno DJ, Herrmann HC, et al. Comparison of two
in elective percutaneous coronary intervention after pretreatment platelet glycoprotein IIb/IIIa inhibitors, tirofiban and abciximab,
with clopidogrel. N Engl J Med 2004; 350(3):232–238. for the prevention of ischemic events with percutaneous coro-
45. Kastrati A, Mehilli J, Neumann FJ, et al. Abciximab in patients nary revascularization. N Engl J Med 2001; 344(25):1888–1894.
with acute coronary syndromes undergoing percutaneous coro- 62. Use of a monoclonal antibody directed against the platelet
nary intervention after clopidogrel pretreatment: the ISAR- glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty.
REACT 2 randomized trial. JAMA 2006; 295(13):1531–1538. The EPIC Investigation. N Engl J Med 1994; 330(14):956–961.
46. Gurbel PA, Bliden KP, Hiatt BL, et al. Clopidogrel for coronary 63. Platelet glycoprotein IIb/IIIa receptor blockade and low-dose
stenting: response variability, drug resistance, and the effect heparin during percutaneous coronary revascularization. The
of pretreatment platelet reactivity. Circulation 2003; 107(23): EPILOG Investigators. N Engl J Med 1997; 336(24):1689–1696.
2908–2913. 64. Randomised placebo-controlled and balloon-angioplasty-con-
47. Savi P, Combalbert J, Gaich C, et al. The antiaggregating activity trolled trial to assess safety of coronary stenting with use of platelet
of clopidogrel is due to a metabolic activation by the hepatic glycoprotein-IIb/IIIa blockade. Lancet 1998; 352(9122):87–92.
cytochrome P450-1A. Thromb Haemost 1994; 72(2):313–317. 65. Montalescot G, Antoniucci D, Kastrati A, et al. Abciximab in
48. Lotfi A, Schweiger MJ, Giugliano GR, et al. High-dose atorvastatin primary coronary stenting of ST-elevation myocardial infarction:
does not negatively influence clinical outcomes among clopidog- a European meta-analysis on individual patients’ data with long-
rel treated acute coronary syndrome patients–a Pravastatin or term follow-up. Eur Heart J 2007; 28(4):443–449.
Atorvastatin Evaluation and Infection Therapy-Thrombolysis in 66. Barrett JS, Murphy G, Peerlinck K, et al. Pharmacokinetics and
Myocardial Infarction 22 (PROVE IT-TIMI 22) analysis. Am Heart pharmacodynamics of MK-383, a selective non-peptide platelet
J 2008; 155(5):954–958. glycoprotein-IIb/IIIa receptor antagonist, in healthy men. Clin
49. Saw J, Brennan DM, Steinhubl SR, et al. Lack of evidence of a Pharmacol Ther 1994; 56(4):377–388.
clopidogrel-statin interaction in the CHARISMA trial. J Am Coll 67. Comparative 30-day economic and clinical outcomes of platelet
Cardiol 2007; 50(4):291–295. glycoprotein IIb/IIIa inhibitor use during elective percutaneous
50. Saw J, Steinhubl SR, Berger PB, et al. Lack of adverse clopidogrel- coronary intervention: Prairie ReoPro versus Integrilin Cost
atorvastatin clinical interaction from secondary analysis of a Evaluation (PRICE) trial. Am Heart J 2001; 141(3):402–409.
randomized, placebo-controlled clopidogrel trial. Circulation 68. Valgimigli M, Campo G, Percoco G, et al. Comparison of angio-
2003; 108(8):921–924. plasty with infusion of tirofiban or abciximab and with implan-
51. Lau WC, Waskell LA, Watkins PB, et al. Atorvastatin reduces the tation of sirolimus-eluting or uncoated stents for acute
ability of clopidogrel to inhibit platelet aggregation: a new drug- myocardial infarction: the MULTISTRATEGY randomized trial.
drug interaction. Circulation 2003; 107(1):32–37. JAMA 2008; 299(15):1788–1799.
52. Gilard M, Arnaud B, Cornily JC, et al. Influence of omeprazole on 69. Moliterno DJ, Yakubov SJ, DiBattiste PM, et al. Outcomes at 6
the antiplatelet action of clopidogrel associated with aspirin: the months for the direct comparison of tirofiban and abciximab dur-
randomized, double-blind OCLA (omeprazole clopidogrel aspi- ing percutaneous coronary revascularisation with stent placement:
rin) study. J Am Coll Cardiol 2008; 51(3):256–260. the TARGET follow-up study. Lancet 2002; 360(9330):355–360.
53. Bliden KP, Dichiara J, Lawal L, et al. The association of cigarette 70. Inhibition of the platelet glycoprotein IIb/IIIa receptor with tir-
smoking with enhanced platelet inhibition by clopidogrel. J Am ofiban in unstable angina and non-Q-wave myocardial infarction.
Coll Cardiol 2008; 52(7):531–533. Platelet Receptor Inhibition in Ischemic Syndrome Management in
54. Wiviott SD, Trenk D, Frelinger AL, et al. Prasugrel compared Patients Limited by Unstable Signs and Symptoms (PRISM-PLUS)
with high loading- and maintenance-dose clopidogrel in patients Study Investigators. N Engl J Med 1998; 338(21):1488–1497.
with planned percutaneous coronary intervention: the prasugrel 71. Hamm CW, Heeschen C, Goldmann B, et al. Benefit of abciximab
in comparison to clopidogrel for inhibition of platelet activation in patients with refractory unstable angina in relation to serum
and aggregation-thrombolysis in myocardial infarction 44 trial. troponin T levels. c7E3 Fab Antiplatelet Therapy in Unstable
Circulation 2007; 116(25):2923–2932. Refractory Angina (CAPTURE) Study Investigators. N Engl J
55. Tcheng JE, Ellis SG, George BS, et al. Pharmacodynamics of Med 1999; 340(21):1623–1629.
chimeric glycoprotein IIb/IIIa integrin antiplatelet antibody 72. Giugliano RP, White JA, Bode C, et al. Early versus delayed,
Fab 7E3 in high-risk coronary angioplasty. Circulation 1994; provisional eptifibatide in acute coronary syndromes. N Engl J
90(4):1757–1764. Med 2009; 360(21):2176–2190.
56. Heeschen C, Hamm CW, Goldmann B, et al. Troponin concen- 73. Montalescot G, Barragan P, Wittenberg O, et al. Platelet glyco-
trations for stratification of patients with acute coronary syn- protein IIb/IIIa inhibition with coronary stenting for acute myo-
dromes in relation to therapeutic efficacy of tirofiban. PRISM cardial infarction. N Engl J Med 2001; 344(25):1895–1903.
Study Investigators. Platelet Receptor Inhibition in Ischemic 74. Maioli M, Bellandi F, Leoncini M, et al. Randomized early versus
Syndrome Management. Lancet 1999; 354(9192):1757–1762. late abciximab in acute myocardial infarction treated with pri-
57. A comparison of aspirin plus tirofiban with aspirin plus heparin mary coronary intervention (RELAx-AMI Trial). J Am Coll
for unstable angina. Platelet Receptor Inhibition in Ischemic Cardiol 2007; 49(14):1517–1524.
Syndrome Management (PRISM) Study Investigators. N Engl J 75. Ellis SG, Tendera M, de Belder MA, et al. Facilitated PCI in
Med 1998; 338(21):1498–1505. patients with ST-elevation myocardial infarction. N Engl J Med
58. Randomised placebo-controlled trial of abciximab before and 2008; 358(21):2205–2217.
during coronary intervention in refractory unstable angina: the 76. Van’t Hof AW, Ten Berg J, Heestermans T, et al. Prehospital
CAPTURE study. Lancet 1997; 349(9063):1429–1435. initiation of tirofiban in patients with ST-elevation myocardial
59. Inhibition of platelet glycoprotein IIb/IIIa with eptifibatide in infarction undergoing primary angioplasty (On-TIME 2): a multi-
patients with acute coronary syndromes. The PURSUIT trial centre, double-blind, randomised controlled trial. Lancet 2008;
investigators. Platelet glycoprotein IIb/IIIa in unstable angina: 372(9638):537–546.
receptor suppression using integrilin therapy. N Engl J Med 77. Boersma E, Harrington RA, Moliterno DJ, et al. Platelet
1998; 339(7):436–443. glycoprotein IIb/IIIa inhibitors in acute coronary syndromes: a
[2/7/010/7:23:1] [53–66]
meta-analysis of all major randomised clinical trials. Lancet 2002; 94. Cohen M, Demers C, Gurfinkel EP, et al. A comparison of low-
359(9302):189–198. molecular-weight heparin with unfractionated heparin for unsta-
78. Lincoff AM, Califf RM, Topol EJ. Platelet glycoprotein IIb/IIIa ble coronary artery disease. Efficacy and Safety of Subcutaneous
receptor blockade in coronary artery disease. J Am Coll Cardiol Enoxaparin in Non-Q-Wave Coronary Events Study Group.
2000; 35(5):1103–1115. N Engl J Med 1997; 337(7):447–452.
79. Cannon CP, Husted S, Harrington RA, et al. Safety, tolerability, 95. Blazing MA, de Lemos JA, White HD, et al. Safety and efficacy
and initial efficacy of AZD6140, the first reversible oral adenosine of enoxaparin vs unfractionated heparin in patients with non-
diphosphate receptor antagonist, compared with clopidogrel, in ST-segment elevation acute coronary syndromes who receive
patients with non-ST-segment elevation acute coronary syn- tirofiban and aspirin: a randomized controlled trial. JAMA 2004;
drome: primary results of the DISPERSE-2 trial. J Am Coll 292(1):55–64.
Cardiol 2007; 50(19):1844–1851. 96. Ferguson JJ, Califf RM, Antman EM, et al. Enoxaparin vs
80. Wallentin L, Becker RC, Budaj A, et al. Ticagrelor versus clopi- unfractionated heparin in high-risk patients with non-ST-
dogrel in patients with acute coronary syndromes. N Engl J Med segment elevation acute coronary syndromes managed with an
2009; 361(11):1045–1057. intended early invasive strategy: primary results of the SYN-
81. Becker RC, Moliterno DJ, Jennings LK, et al. Safety and tolerability ERGY randomized trial. JAMA 2004; 292(1):45–54.
of SCH 530348 in patients undergoing non-urgent percutaneous 97. Martin JL, Fry ET, Sanderink GJ, et al. Reliable anticoagulation
coronary intervention: a randomised, double-blind, placebo-con- with enoxaparin in patients undergoing percutaneous coronary
trolled phase II study. Lancet 2009; 373(9667):919–928. intervention: the pharmacokinetics of enoxaparin in PCI (PEPCI)
82. The Thrombin Receptor Antagonist for Clinical Event Reduction study. Catheter Cardiovasc Interv 2004; 61(2):163–170.
in Acute Coronary Syndrome (TRA*CER) trial: study design and 98. Montalescot G, White HD, Gallo R, et al. Enoxaparin versus
rationale. Am Heart J 2009; 158(3):327–334 e324. unfractionated heparin in elective percutaneous coronary inter-
83. Hirsh J, Warkentin TE, Shaughnessy SG, et al. Heparin and Low- vention. N Engl J Med 2006; 355(10):1006–1017.
Molecular-Weight Heparin Mechanisms of Action, Pharmacoki- 99. Mehta SR, Granger CB, Eikelboom JW, et al. Efficacy and safety
netics, Dosing, Monitoring, Efficacy, and Safety. Chest 2001; 119 of fondaparinux versus enoxaparin in patients with acute coro-
(90010):64S–94S. nary syndromes undergoing percutaneous coronary interven-
84. Ogilby JD, Kopelman HA, Klein LW, et al. Adequate heparin- tion: results from the OASIS-5 trial. J Am Coll Cardiol 2007;
ization during PTCA: assessment using activated clotting times. 50(18):1742–1751.
Cathet Cardiovasc Diagn 1989; 18(4):206–209. 100. Alexander JH, Becker RC, Bhatt DL, et al. Apixaban, an oral,
85. King SB III, Smith SC Jr, Hirshfeld JW Jr, et al. 2007 focused direct, selective factor Xa inhibitor, in combination with antipla-
update of the ACC/AHA/SCAI 2005 guideline update for telet therapy after acute coronary syndrome: results of the
percutaneous coronary intervention: a report of the American apixaban for Prevention of Acute Ischemic and Safety Events
College of Cardiology/American Heart Association Task Force (APPRAISE) trial. Circulation 2009; 119(22):2877–2885.
on Practice guidelines. J Am Coll Cardiol 2008; 51(2):172–209. 101. Matthai WH Jr. Use of argatroban during percutaneous coronary
86. Chew DP, Bhatt DL, Lincoff AM, et al. Defining the optimal interventions in patients with heparin-induced thrombocytope-
activated clotting time during percutaneous coronary interven- nia. Semin Thromb Hemost 1999; 25(suppl 1):57–60.
tion: aggregate results from 6 randomized, controlled trials. 102. Cruz-Gonzalez I, Sanchez-Ledesma M, Baron SJ, et al. Efficacy
Circulation 2001; 103(7):961–966. and safety of argatroban with or without glycoprotein IIb/IIIa
87. Hillegass WB, Brott BC, Chapman GD, et al. Relationship inhibitor in patients with heparin induced thrombocytopenia
between activated clotting time during percutaneous interven- undergoing percutaneous coronary intervention for acute coro-
tion and subsequent bleeding complications. Am Heart J 2002; nary syndrome. J Thromb Thrombolysis 2008; 25(2):214–218.
144(3):501–507. 103. Campbell KR, Mahaffey KW, Lewis BE, et al. Bivalirudin in
88. Brener SJ, Moliterno DJ, Lincoff AM, et al. Relationship between patients with heparin-induced thrombocytopenia undergoing
activated clotting time and ischemic or hemorrhagic complica- percutaneous coronary intervention. J Invasive Cardiol 2000;
tions: analysis of 4 recent randomized clinical trials of percuta- 12(suppl F):14F–19F.
neous coronary intervention. Circulation 2004; 110(8):994–998. 104. Mahaffey KW, Lewis BE, Wildermann NM, et al. The antico-
89. A comparison of reteplase with alteplase for acute myocardial agulant therapy with bivalirudin to assist in the performance of
infarction. The Global Use of Strategies to Open Occluded Cor- percutaneous coronary intervention in patients with heparin-
onary Arteries (GUSTO III) Investigators. N Engl J Med 1997; induced thrombocytopenia (ATBAT) study: main results. J Inva-
337(16):1118–1123. sive Cardiol 2003; 15(11):611–616.
90. Granger CB, Hirsch J, Califf RM, et al. Activated partial throm- 105. Lincoff AM, Bittl JA, Harrington RA, et al. Bivalirudin and
boplastin time and outcome after thrombolytic therapy for acute provisional glycoprotein IIb/IIIa blockade compared with hep-
myocardial infarction: results from the GUSTO-I trial. Circulation arin and planned glycoprotein IIb/IIIa blockade during percuta-
1996; 93(5):870–878. neous coronary intervention: REPLACE-2 randomized trial.
91. Ellis SG, Roubin GS, Wilentz J, et al. Effect of 18- to 24-hour JAMA 2003; 289(7):853–863.
heparin administration for prevention of restenosis after uncom- 106. Stone GW, McLaurin BT, Cox DA, et al. Bivalirudin for patients
plicated coronary angioplasty. Am Heart J 1989; 117(4):777–782. with acute coronary syndromes. N Engl J Med 2006; 355
92. Friedman HZ, Cragg DR, Glazier SM, et al. Randomized pro- (21):2203–2216.
spective evaluation of prolonged versus abbreviated intravenous 107. Stone GW, Witzenbichler B, Guagliumi G, et al. Bivalirudin
heparin therapy after coronary angioplasty. J Am Coll Cardiol during primary PCI in acute myocardial infarction. N Engl J
1994; 24(5):1214–1219. Med 2008; 358(21):2218–2230.
93. Rabah MM, Premmereur J, Graham M, et al. Usefulness of 108. Kastrati A, Neumann FJ, Mehilli J, et al. Bivalirudin versus
intravenous enoxaparin for percutaneous coronary intervention unfractionated heparin during percutaneous coronary interven-
in stable angina pectoris. Am J Cardiol 1999; 84(12):1391–1395. tion. N Engl J Med 2008; 359(7):688–696.
[13/7/010/12:55:54] [67–84]
Vascular access and closure

Zoltan G. Turi
INTRODUCTION ACCESS SITE

The familiar percutaneous vascular access approach pioneered The choice of access site is influenced by the end organ to be
by Seldinger in 1953 (1) revolutionized invasive medicine. It investigated or treated, the size of the catheters to be used, and
provided an alternative to the time-consuming and morbid sur- the success and complication rates associated with each
gical cutdown techniques that had been part of vascular access approach. Most diagnostic and therapeutic procedures in the
since before the time of Werner Forssmann’s self-catheterization United States utilize femoral access, in contrast to other parts of
by cutdown in 1929. Cardiac catheterization, however, was still the world where coronary procedures are now most commonly
primarily via surgical access through the 1960s, with brachial performed via the radial approach. The brachial approach is a
cutdowns a ubiquitous feature of the Sones technique (2). Periph- distant third in the terms of frequency, followed by a number of
eral angiography and intervention were also largely by femoral relatively uncommon or rare arterial access routes, including
cutdown, with a high associated access site complication rate (3). retrograde popliteal, axillary, translumbar aortic, direct carotid,
The introduction of the Judkins technique in the 1960s (4) was direct left ventricular apex, and direct infrapopliteal access.
associated with widespread adoption of Seldinger’s method. A number of randomized studies have compared access
The radial approach, nearly universally by percutaneous punc- routes, most commonly transfemoral and transradial access,
ture, was introduced by Campeau for cardiac catheterization in while one has compared transradial, transfemoral, and trans-
1989 (5) and by Kiemeneij for cardiac intervention in 1993 (6). brachial (8). Meta-analyses in general have convincingly dem-
Percutaneous brachial access, although occasionally utilized onstrated the highest success rates with the femoral approach,
for peripheral angiography starting in the 1950s (7), now has and the lowest complication rates with radial access (11), with
limited overall utilization, and the least appealing overall safety intermediate success rate but the highest complication rate with
profile (8). percutaneous transbrachial procedures (8).
The Seldinger method has been subjected to relatively
few investigations and has changed remarkably little over a
half century. Two major modifications are noteworthy: first, the
Access Technique
Virtually all percutaneous access now involves anterior wall
original Seldinger technique involved through-and-through
hollow needle puncture of the artery followed by introduction
puncture of the artery, typically with a stylet in the needle,
of a guidewire through the needle, after which a catheter
and pullback with the stylet removed until flashback was
(typically a sheath) is introduced over the guidewire (Fig. 8.1).
noted. Because the back wall puncture facilitates bleeding
An uncommon variation in cardiac catheterization, but com-
and hence hematoma formation, and has been implicated as a
monly used for intravenous lines, is the insertion of a sleeve over
cause of retroperitoneal hemorrhage (RPH), the “modified
the outside of the needle, a technique that is occasionally used for
Seldinger technique” has now been adopted almost univer-
radial catheterization as well. The use of sheaths, typically with
sally. This modification involves puncture of the anterior wall
sidearms for pressure monitoring, blood sampling, or angiog-
only (Fig. 8.1). The second major innovation was the introduc-
raphy requires an approximately 0.7 mm increase in the size of
tion of sheath introducers in 1979 (9); this eliminated the
the arteriotomy. A common misconception is that a 6 French (F)
introduction of multiple catheters directly through the skin,
sheath is 2 mm in size (one French size is the equivalent of
obviated the need for performing exchange procedures every
0.33 mm); although a 6F sheath will accommodate a 2-mm
time another catheter was utilized, and resulted in a smoother
catheter, it has an outer diameter in the range of 2.7 mm. Thus,
arteriotomy without the fraying at the edges that was common
by convention, catheter size refers to outer diameter, whereas
after multiple catheter introductions. It also prevented accu-
sheath size refers to inner lumen; in fact a catheter equivalent in
mulation of debris in the tip of the catheter as it passed through
outer diameter to a 6F sheath would be 8F in size.
skin, muscle and cartilage. Although debris embolization from
catheter tips remains a source of morbidity, it is now rarely
associated with naked catheter passage through the skin (10). Access Needle Size
Vascular access has been subjected to significant clinical Most cardiologists continue to utilize large bore 18-gauge nee-
investigation in the past decade and an evidence base is dles for percutaneous vascular access. These provide consider-
accumulating that addresses both access and closure methods. able “splash-back” of blood on arterial entry. In contrast, most
Although incomplete, the literature now has a number of radiologists use micropuncture techniques, as do cardiologists
important studies addressing access site selection, radial versus performing radial access, typically with a 21-gauge needle,
femoral access, puncture location, adjunctive use of fluoros- which is approximately 0.8 mm in diameter. Micropuncture
copy and ultrasound, and manual compression versus device- results in only a trickle of blood through the needle at the time of
based vascular closure; all will be covered in this chapter. vessel entry, which is a significant potential safety advantage,
[13/7/010/12:55:54] [67–84]
Figure 8.1 The modified Seldinger technique. A hollow needle punctures the anterior wall of the artery only, with splashback noted. A
guidewire is advanced through the needle, followed by a dilator and sheath. The final image shows the sheath remaining in the artery after the
dilator is withdrawn. Source: Courtesy of Dr Edward Perper, AnimationMD.
although arterial entry is less obvious. In contrast, the 18-gauge potential for significantly greater extravasation of blood, leading
needle is approximately 1.3 mm in outer diameter, 56% larger in turn to higher risk of hemorrhagic complications. Although
than the 21-gauge needle, and results in nearly six times as intuitively appealing, there is as yet no compelling evidence
much blood flow (Fig. 8.2). Thus, errant vessel puncture has the base for the benefits of micropuncture access.
FEMORAL ACCESS
Puncture Location
The femoral artery continues to dominate access in the United
States because of two advantages: first, large devices, 24F or
greater, can be accommodated if vessels are both large and
compliant, and second, relative ease of access. Three primary
landmarks have been utilized for arterial access: the inguinal
crease, the point of maximal pulsation, and a line drawn between
the anterior superior iliac crest and the symphysis pubis. The
most common landmark continues to be the inguinal crease, an
unfortunate choice since it tends to lead to puncture outside the
common femoral artery. In a study by Grier et al. (13) surveying
the practices of 200 interventional radiologists and cardiologists,
40% used the inguinal crease as the sole landmark for initial
puncture. Figure 8.3 shows the common femoral artery as tradi-
tionally depicted in most textbooks. There are three major mis-
conceptions represented in this cartoon. First, the inguinal
ligament is depicted above the femoral head, a relationship that
is less common than most operators realize. Second, the inguinal
Figure 8.2 The advantage of micropuncture compared with con-
crease is depicted as overlying the center of the femoral head.
ventional access showing the influence of needle size on blood flow.
The inset shows a standard gauge and micropuncture needle for Finally, the inguinal crease is depicted as being superior to the
comparison. Because the 18-gauge needle is 56% larger than a femoral bifurcation. In fact, the inguinal ligament descends over
micropuncture needle, applying Ohm’s and Poiseuille’s laws results the upper half of the femoral head in a significant percent of
in a theoretical nearly sixfold increase in blood flow if there is an cases, and the crease is an average 6 mm below the femoral
errant puncture. Source: Adapted from Ref. 12. bifurcation (13), which in turn is at or below the femoral head in
more than three-quarters of patients (14).
[13/7/010/12:55:54] [67–84]
VASCULAR ACCESS AND CLOSURE 69
Figure 8.3 Cartoon on the left showing anatomic relationship of femoral vasculature, bony landmarks and inguinal ligament. Cineangiogram
of the femoral artery corresponding to the cartoon is on the right. There are multiple, almost universal misconceptions in the image on the left.
First, the inguinal ligament is portrayed as cranial to the femoral head, an unwarranted assumption in many patients. The inguinal crease is
portrayed as being over the center of the femoral head and cranial to the femoral bifurcation, neither of which corresponds to the evidence
base. Arrows on right point to conventional, but inaccurate, borders for safe access, the top of the femoral head to the bifurcation of the
femoral artery. Puncture near either arrow in the figure on the right is associated with an increased risk of complications. The higher arrow is
above the inferior epigastric artery, and the risk of retroperitoneal hemorrhage would be increased if puncture were above the “x” in the image.
Puncture near the lower arrow, which is at the femoral bifurcation, would be substantially below the femoral head, increasing the risk of
hematoma, pseudoaneurysm and arteriovenous fistula, even if needle entry were in the common femoral artery. Abbreviations: CFA,
common femoral artery; PFA, profunda femoris artery; SFA, superficial femoral artery. Source: For figure on left: From Ref. 15.
Puncture at the inguinal crease tends to predispose to low downward course toward the inguinal ligament. It approaches
sticks, especially in obese patients, including entry into one of but does not cross the ligament and then ascends cranially
the femoral bifurcation vessels, either the superficial femoral toward the epigastrium, where its circulation overlaps with
artery or the profunda femoris. Low puncture (below the that of the internal mammary arteries. Punctures below the
femoral head), whether entry into the common femoral or most inferior sweep of the IEA are least likely to be associated
one of its branches, predisposes to several complications, with RPH. Typically the IEA does not descend below the
most notably arterial pseudoaneurysm (16). Postprocedure femoral head centerline.
compression of the artery with catheter/sheath removal is The location of the femoral bifurcation, on the basis of a
impaired by lack of the anvil represented by the femoral 200-patient study (20), is at or below the bottom of the femoral
head itself; compression against layers of fat, muscle and head in approximately 77% of patients and above the centerline
other soft tissue elements can result in inadequate hemostasis of the femoral head in less than 2% of cases. Because of the
and subsequent pseudoaneurysm formation (Fig. 8.4). Two greater risk associated with high puncture, and the fact that the
other consequences of note accompany low puncture: the exact excursion of the IEA is variable among patients and is
venous circulation runs closely parallel or overlaps with the unknown in most at the time of access, an optimal location,
femoral bifurcation vessels, increasing the risk of arteriovenous below the centerline of the femoral head but high enough to
fistulae (17). Femoral nerve branches approximate the artery at minimize the risk of entry into the femoral bifurcation vessels,
and below the femoral bifurcation, increasing the risk of nerve is shown in Figure 8.5. Puncture 5 mm or more below the
compression with hematoma formation or, if vascular closure centerline of the femoral head results in access below the IEA in
devices (VCDs) are used, of injury to the nerve from direct 97% of patients (21).
trauma, including rare cases of stitch or clip strangulation.
Of far more concern in terms of overall morbidity and Fluoroscopy and Ultrasound Guidance
mortality, however, is high puncture, associated with up to an Fluoroscopy to guide common femoral artery access was widely
18:1 odds ratio for RPH (18). Once considered to be a problem adopted by radiologists thirty years ago (22) (Fig. 8.6A). Local-
primarily with punctures above the femoral head, an increasing izing the relatively small target zone can be accomplished by
evidence base has explored the relative location of femoral placing a hemostat at the bottom edge of the femoral head as
punctures with relationship to various bony and vascular originally described by Kim (16) or over the target area below
landmarks. Figure 8.5 shows the relative position of the ingui- the femoral head centerline. An important additional element of
nal ligament to various anatomical features. The most useful the technique is iterative fluoroscopy once the needle has
single landmark is the inferior epigastric artery (IEA); this entered the tissue track but before the needle is advanced far
vessel arises from the external iliac artery and initially has a enough to enter the artery (Fig. 8.6B). Removing his or her hands
[13/7/010/12:55:54] [67–84]
Figure 8.4 (See color insert) Computed tomographic angiogram

showing relationship of the iliac and common femoral arteries to the
bony structures of the pelvis and the femoral head. Note that the
distance from the artery to a bony support structure (that can serve
as an anvil against which compression forces can be applied) is
shortest (green arrow) over the common femoral artery. The lower
red arrow shows the increased distance between the shaft of the
femur and the distal common femoral artery if punctures are below
the femoral head; this requires compression through additional soft
tissue including muscle, fat and cartilage with associated increased
risk of pseudoaneurysm. Puncture into the superficial femoral artery
is associated with increased risk of arteriovenous fistulae. The upper
red arrow shows the extensive distance between the external iliac
artery, which descends toward the retroperitoneum, and the deep
pelvis; high puncture has odds ratios up to 18 to 1 for RPH in fully
anticoagulated patients (18). Abbreviation: RPH, retroperitoneal
hemorrhage.
Figure 8.5 (A) Anatomy for femoral puncture. A, bottom of femoral

from the field, the operator confirms that the needle tip is close head; B, centerline of femoral head; C, line drawn from anterior
the desired target area, and then advances into the anterior wall superior iliac crest to symphysis pubis; D, inguinal ligament;
of the artery. This additional step shows the needle track to be E, inferiormost excursion of the inferior epigastric artery; and
outside the desired horizontal or vertical plane in a large per- F, inguinal crease. The oval depicts the ideal location for femoral
centage of cases, and requires repositioning prior to vessel entry. access. (B) The landmarks depicted in (A) are seen here super-
Even so, fluoroscopy without this additional step has been imposed on a femoral angiogram. Landmarks A through C can be
shown to decrease overall complication rates in one retrospec- detected on plain fluoroscopy. The sheath is seen to enter in the
tive analysis by Fitts and colleagues (23), which demonstrated ideal target location. Source: From Ref. 19.
lower rates of pseudoaneurysm and arterial injury in the prac-
tices of operators who used fluoroscopic guidance. A recent
randomized prospective comparison of fluoroscopic versus tra-
ditional landmark guided femoral access by Abu-Fadel et al. An alternative approach is utilized by many interven-
(24) demonstrated fewer punctures below the femoral head with tional radiologists, and involves ultrasound-guided access, typ-
fluoroscopy but was underpowered to assess complication rates. ically with visualization of the artery in cross sectional view
Importantly, this study did not utilize the iterative fluoroscopic (26). This technique allows predictable arterial entry and min-
technique described above and had nearly half the punctures imizes the risk of inadvertent venous puncture. It does not
above the centerline of the femoral head, an undesirable conse- provide similar guidance as fluoroscopy with regard to avoid-
quence of reliance on hemostat fluoroscopy as the sole source of ing high puncture, and preliminary results of the Femoral
radiographic guidance. Over all, the potential benefit remains Arterial Access with Ultrasound (FAUST) trial (27) included a
unproven (25). substantial number of punctures above the inguinal ligament.
[13/7/010/12:55:54] [67–84]
Figure 8.6 (A) Hemostat over the femoral head, placed over the ideal puncture location at the beginning of the femoral access procedure.
(B) The iterative fluoroscopy technique, showing repeat fluoroscopy after the needle has traverse to a depth just prior to arterial entry.
Typically, a pulsation can be felt through the needle at this point. Note the needle tip (dashed arrow) in the ideal location: just below the
centerline (dashed line) and over the medial portion of the femoral head.
The technique requires modest additional preparation in the typical angle of puncture is approximately 458, and should be
cath lab, does appear to decrease the number of needle passes aligned in the same plane as the artery. Using the iterative
required to access the artery, and reduces inadvertent venous fluoroscopic technique, flashback of blood should indicate
access, but has not been adopted by many cardiologists. passage through the anterior wall of the artery. The typical
Finally, techniques such as percutaneous aortic valve J-tipped guidewire can be seen to traverse the iliac circulation
interventions, percutaneous ventricular assist devices, and to the aorta (patient’s left side), and fluoroscopy is usually not
abdominal aortic aneurysm stent-graft placement have necessary when a 0.035@ J-tipped guidewire passes freely.
increased the importance of first pass access into the femoral However, if a micropuncture technique is used, the micro-
artery. Besides the fluoroscopy and ultrasound methods puncture wire must be visualized prior to sheath introduction,
already described, techniques include contralateral catheter- since the wire can easily enter a small caliber branch vessel, and
ization with contrast media injection, or placement of a pigtail pushing a sheath into small vessels such as the lateral circum-
catheter into the common femoral artery to provide a target for flex of the hip can result in arterial rupture.
the needle on fluoroscopy. An adjunctive technology that is occasionally used is the
SmartNeedle (Escalon, New Berlin, Wisconsin). This uses Dop-
Femoral Access Technique pler technology through an obturator in the needle to localize
Once the target location is determined, local anesthesia is infil- venous or arterial pulsations. If the groin is scarred, as is the
trated several centimeters below the target zone. Lidocaine is case after multiple prior catheterizations, a stiff 0.035@ J-tipped
typically used for local anesthesia; because of zero order kinetics, guidewire should be placed through the needle to provide
the percentage of lidocaine in the injectate determines duration of sufficient support for sheath entry, or, if micropuncture is
anesthesia and not level of anesthesia. Thus for procedures that used, a reinforced micropuncture sheath can be utilized. If
are relatively short in duration, 1% local anesthetic is adequate. there is extensive atherosclerotic disease or tortuosity in the
For longer procedures, 2% lidocaine may be preferable; an alter- external or common iliac, use of a hydrophilic guidewire can be
native, utilized for long duration procedures such as ablations, is considered, although care should be taken to avoid advancing
bupivacaine. Importantly, each percent solution means 10 mg of into sidebranches or creating an intimal dissection, and with-
lidocaine injected subcutaneously; thus, 10 to 20 cc of injectate will drawal of a hydrophilic guidewire through an introducer nee-
result in systemic levels of lidocaine, and in older patients, those dle can readily lead to severing the guidewire tip. On occasion,
with small body mass or those with hepatic failure, lidocaine femoral access is desirable in a patient with aorto- or iliofe-
toxicity should be considered in cases of altered mental status or moral bypass grafts. There is no contraindication to access in
seizures. Some laboratories use lidocaine with epinephrine to this setting, but automated compression devices should not be
decrease the rate of absorption, thereby increasing duration of used for hemostasis.
anesthesia, and decreasing oozing in the tissue track. Buffering Anticoagulation is rarely used for routine diagnostic
the solution with sodium bicarbonate decreases the pain associ- angiography via the femoral route. Early in the history of
ated with local anesthesia (28). catheterization, prolonged catheter dwell times, thrombogenic
The access needle is approximately 7 cm in length; the materials, and the need for prolonged guidewire placement
depth of access will depend on the patient’s body mass. The while catheters were exchanged all predisposed to thrombus
[13/7/010/12:55:54] [67–84]
Figure 8.7 Femoral angiography in (A) contralateral and (B) ipsilateral views. The solid arrow shows the point of sheath entry. The circle
overlies the femoral bifurcation. The dashed arrow denotes the inferiormost excursion of the inferior epigastric artery, which descends to, but
does not cross the inguinal ligament.
formation. When intervention is planned, we advocate routine Complications

femoral angiography prior to anticoagulation as described below. The precise complication rate of femoral artery access is elusive,
and predicated on wide variability of definitions, in particular
hematoma rates. The definitions for significant hematoma vary
Femoral Angiography from 2 to 15 cm in the literature, and rates range from less than
Angiography of the femoral artery after sheath placement 1% to more than 10%. Using a 10-cm threshold, a 12,000 patient
allows the operator to assess three parameters important to study from the Montreal Heart Institute (30) found a 1.8%
avoidance of complications: first, location of puncture; second, vascular complication rate associated with femoral diagnostic
size of the artery; and third, presence of atherosclerotic disease catheterization and 4.0% associated with interventions.
at the puncture site. Most importantly, if angiography is While the most common complication over all is hematoma
performed immediately after sheath access, in the presence of formation, the most lethal is RPH, with a mortality in the range
anatomy that predefines high risk for postprocedure vascular of 5% (18,29). The incidence of RPH has increased substantially
complications (e.g., puncture above the inguinal ligament), in the interventional era, and approaches 1% (18,29,31). Pseu-
anticoagulation and aggressive antiplatelet therapy can be doaneurysm occurs in a range from 1% to 5% (the higher values
avoided, since the complication rate is substantially higher in are seen with routine ultrasound that detects clinically insig-
those settings (29). Angiography is typically performed in the nificant pseudoaneurysms) (32), while arteriovenous fistulae,
ipsilateral view (Fig. 8.7); this optimizes visualization of the significant vessel dissection, obstruction, neural damage, and
femoral bifurcation, although the sheath entry point is fre- venous obstruction occur in less than 1% of cases. The other
quently obscured. Placing tension on the sheath and moving clinically important complication, infection at the access site,
it side to side during contrast injections can help localization. occurs in less than 0.3% of cases, but has a mortality in the
Alternatively, an ipsilateral caudal or contralateral view can be 6% range (33).
used. There is compelling evidence that bleeding episodes are a
Although the mean size of the common femoral artery is strong independent predictor of adverse outcomes (34). In
6 to 8 mm, the minimal size of the vessel is only 5.0 mm in general bleeding complications have decreased substantially
women and 6.3 mm in men (20). At least one-quarter of women in the past two decades. A comparison (35) of three major
have minimal lumen size of 4 mm or less; in this setting abciximab trials revealed a greater than 10% rate of major
the external diameter of a 6F sheath will occlude three-quarters bleeding including transfusions in the EPIC trial (36), decreas-
or more of an otherwise healthy artery. In addition to ing over the subsequent four years to approximately 2% in the
women, diabetics also have smaller vessel diameters. Various EPISTENT study (37). A number of factors appear to have led
interventions are simply incompatible with common femoral to the decline, including lower heparin dosage, weight adjusted
artery size in patients with small arteries, in particular abdomi- heparin, elimination of postprocedure anticoagulation, and
nal aortic aneurysm stent-grafts and percutaneous aortic valve smaller sheath sizes. A further decline in the bleeding rate,
replacement that require sheaths that are 8 mm or greater in and an overall decline in vascular complications, has continued
diameter. throughout the past decade (38), reflecting more prudent
[13/7/010/12:55:54] [67–84]
Figure 8.8 Cartoon showing relationship between anatomy and deployment of vascular closure devices, as seen from a left lateral view of
the patient. The common femoral artery is seen over the femoral head, with deployment of various closure devices depicted. As the artery
descends more proximally into the pelvis, layers of fascia and muscle are interposed between the skin surface and the arterial wall. These
interfere with delivery of the plug, clip, or suture onto the arterial surface, and increase the risk of retroperitoneal hemorrhage. Source: From
Ref. 19.
anticoagulation and antiplatelet regimens, and better vascular operator should consider allowing anticoagulation to wear off,
access and closure techniques. and pulling the sheath manually. Although vascular compres-
The relatively high RPH rates, despite an overall decline sion may be ineffective, the rate of RPH is substantially reduced
in hemorrhagic complications, appears related to glycoprotein in the absence of anticoagulation (39). Finally, if the patient is
IIb/IIIa use (39) as well as vascular closure devices (VCDs) (18). actively hemorrhaging, contralateral access and balloon occlu-
In part because manual compression is performed only after sion of the puncture site can be life saving and in most cases
the activated clotting time has normalized, the rate of RPH, therapeutic, sometimes with the incorporation of a fabric covered
even after intervention, is relatively low when manual com- stent if balloon tamponade alone is inadequate (41). In general, all
pression is used (39); RPH rarely occurs in unanticoagulated hypotension after femoral access should raise the question of
patients. In contrast, VCDs are deployed when patients are RPH, since hypotension is the most sensitive early marker.
fully anticoagulated, and the rate of RPH is dramatically Although a number of other signs and symptoms are more
increased (18). The problem is exacerbated by the interposition specific (29), the time course of RPH frequently does not allow
of soft tissue, especially muscle and other connective tissue, for delay in diagnosis. CT scanning is diagnostic, but in the
between the skin surface and the external iliac artery as it dives setting of hemodynamic instability, a prompt return to the
toward the retroperitoneum, resulting in inability to advance the catheterization laboratory for diagnosis and intervention is pref-
plug/clip/staple/knot or other closure device onto the surface erable, assuming that trained staff and appropriate equipment are
of the artery, even though an anchor is in place inside the artery available; if not, surgery should not be delayed. Anticoagulation
(Fig. 8.8). Although the evidence base is not conclusive (40), it is should be reversed in this setting and early blood transfusion can
generally advisable to avoid VCD use with known high femoral be life saving.
punctures. Similar to RPH, vascular access site infection is associated
Several algorithms for management of high puncture with VCD use. This represents one of the worst complications
have been suggested. First, in the setting of elective catheter- of femoral access, although relatively rare, occurring in as few
ization, anticoagulation and antiplatelet therapy are best as 0.25% of cases (33). It occurs a median of eight days after
avoided if a high stick occurs. Diagnostic catheterization can access (range two days to approximately one month), is typi-
proceed, but the patient should be brought back for any inter- cally caused by Staphylococcus aureus, is frequently blood cul-
vention. Second, if the patient is already anticoagulated, the ture positive and occurs primarily in diabetics and the immune
[13/7/010/12:55:54] [67–84]
suppressed. A mycotic pseudoaneurysm is present in nearly body (Perclose; StarClose), temporary foreign body (Angio-Seal,
half of cases, and if so, surgical exploration is required. Aggres- Duett, Mynx), or no foreign body at all (Catalyst, Cardiva
sive, early and parenteral antibiotics are usually needed. A high Medical, Sunnyvale, California).
index of suspicion is important. In practice, even in the absence In general, VCDs facilitate closure by decreasing time to
of fever or other signs of infection, the occurrence of late hemostasis and time to ambulation. This is potentially offset by
bleeding or oozing at the access site should always raise the cost and complications. Failure to obtain hemostasis is the
suspicion of dehiscence secondary to localized infection. primary manifestation of device failure, typically a minor prob-
Pseudoaneurysms are now readily diagnosable with lem after diagnostic catheterization, but potentially a cause of
ultrasound, and can be treated conservatively in most cases if major morbidity if deployment is attempted while patients are
less than 2 cm in diameter. The primary causes are low punc- still fully anticoagulated after interventional procedures. Pas-
ture as already discussed and inadequate compression time. sive closure devices, typically unanchored plugs, have what
Although ultrasound-guided compression and occasionally appears to be the highest failure rate (45). There is a significant
surgery are required, both procedures are associated with learning curve incorporated in the use of VCDs (46,47).
significant discomfort, and most pseudoaneurysms are now An important application of VCDs has been preclosure, a
treated with offlabel ultrasound-guided thrombin injection technique that involves deploying sutures around the arterio-
with a high rate of success and low complication rate (42,43). tomy without tying a knot until sheath withdrawal, and is
Arterial obstruction is rare, and usually occurs in the performed prior to placing a large lumen sheath. This tech-
presence of preexisting atherosclerotic disease; some athero- nique has been used with a high degree of success to close
sclerosis at the access site is relatively common (20); occlusion arterial fenestration after withdrawal of large sheaths up to 28F
secondary to VCDs is the most common scenario (44). Dissec- used for a variety of arterial and to a lesser degree venous
tion at the femoral access site is typically retrograde, thus the interventions, avoiding the need for surgical access or closure
entry point of the false channel faces opposite to the direction of (48).
blood flow. These are usually self-limited and rarely require The cost-to-benefit ratio is the subject of some contro-
intervention as long as the operator does not try to advance versy; the most compelling study demonstrating reduced costs
guidewires or catheters aggressively into the false lumen. with VCDs was predicated on a lower complication rate than
Nerve compression typically manifests as dysesthesia radiating with manual compression along with earlier discharge that
to the medial aspect of the knee and occasionally to the ankle. resulted in decreased resource utilization (49). Two meta-anal-
This is typically secondary to compression of a femoral nerve yses concluded that complication rates associated with VCDs
branch by hematoma, and usually resolves within days to were higher than those associated with manual compression
weeks. Direct injection of local anesthetic into the nerve or (50,51), but the quality of the underlying studies in these meta-
transsection by the needle or sheath along with strangulation of analyses was poor, and significant study design issues, includ-
the nerve by a VCD can all theoretically cause long-term ing incorporation of learning curves and variable device plat-
sequelae, but these complications are rare. forms, have made broad conclusions unreliable. Two large
propensity analyses have been promising, and suggest that
with proper patient selection, meticulous access technique,
and experienced operators, VCDs can in fact demonstrate
Vascular Closure Devices equivalent or lower complication rates than manual compres-
Vascular closure of percutaneous access sites was invariably by sion (52,53) (Fig. 8.10). Nevertheless, a review of the FDA
manual or mechanical device compression for the first 40 years database confirms that some complications of VCDs are addi-
of the Seldinger technique. A variety of compression devices tive to those seen with manual compression, in particular RPH,
were introduced as early as the 1950s and vary in design from infections, and occasional vascular occlusion. The ultimate
mechanical clamps to inflatable bladders such as the FemoStop decision regarding risk-to-benefit ratios remains with individ-
(St. Jude Medical, St. Paul, Minnesota, U.S.). Devices designed ual operators, without a consensus from the evidence base.
to facilitate hemostasis can be classified into a number of
subcategories (Fig. 8.9). In general, they can be thought of as
invasive if they are deployed in the tissue track itself, as
opposed to various noninvasive agents, such as topical hemo- RADIAL ACCESS
static patches, that are applied to the surface only. A second Anatomy
subclassification considers devices that suture (Perclose, Abbott The circulation to the hand is redundant in over 95% of
Medical, Santa Clara, California, U.S.), clip (StarClose, Abbott patients, with a loop between the ulnar and radial arteries
Medical), staple (AngioLink, Medtronic, Minneapolis, Minnesota, providing crossover filling. Because the radial artery is smaller
U.S.), or sandwich (Angio-Seal, St. Jude Medical, St. Paul, in size than the common femoral, there is some limitation in the
Minnesota) the arteriotomy between an internal anchor and range of interventional procedures that can be performed via
an external plug; these are active approximators. Passive approx- radial access. The learning curve is significant, procedure times
imators place a device on top of the arteriotomy without an and radiation to the patient and operator are potential issues,
anchor left in place (Mynx, Access Closure, Mountain View, and spasm of the arm vessels is occasionally painful and limits
California). A third subclassification relates to the use of hemo- applicability of the procedure. Nevertheless, radial catheter-
static agents such as collagen or thrombin (Angio-Seal; Duett, ization is now the dominant access route in some parts of the
Vascular Solutions, Minneapolis, Minnesota) or sealing agents world, and its use is increasing steadily in the United States as
such as polyethylene glycol (Mynx) or polyglycolic acid (Exo- well (54).
Seal, Ethicon, Warren, New Jersey, U.S., not FDA approved), The technique requires confirming the redundancy of
typically as a plug in the tissue track. Finally, devices can be radial and ulnar circulation. The standard approach is the
subclassified by whether they leave behind a permanent foreign Allen test (55), with occlusive pressure applied to both the
[13/7/010/12:55:54] [67–84]
Figure 8.9 A gallery of vascular closure devices. The top row (A–C) features devices that provide for approximation of the puncture site in the
arterial wall by creating a “sandwich” of anchor and collagen (Angio-Seal), suturing (Perclose), or clipping (StarClose) the fenestration closed.
The second row (D–F) features passive closure devices that seal the arteriotomy by placing sealing plugs (Mynx, ExoSeal) in the tissue track on
top of the hole in the artery created by the puncture. The Catalyst (F) uses compression from inside the artery to achieve hemostasis, after which
the nitinol disk is collapsed and withdrawn. The final two rows feature two devices that start the closure process at the time of access: Arstasis
(Arstasis, San Carlos, California, U.S.) (G) is inserted by placing a small device using micropuncture to establish access to the vessel (small
access track—arrow); the device in turn deploys a needle that establishes a diagonal track across the arterial wall (double arrow) through which
the procedure sheath is placed. After withdrawal (H), hydrostatic pressure of blood against the arterial wall (solid arrows) contributes to sealing
the arteriotomy. The Femoral Introducer Sheath and Hemostasis (FISH) device (Morris Innovative Research, Bloomington, Indiana, U.S.)
introduces a biodegradable material (small intestinal submucosa) wrapped around the sheath at the time of access (I), which forms a plug (J) at
the time of sheath withdrawal. The ExoSeal device is not FDA approved. See text for manufacturers of devices in panels (A) to (F).
[13/7/010/12:55:54] [67–84]
Figure 8.10 Four analyses comparing manual compres-

sion with VCD complication rates. The studies by Koreny
(50) and Nikolsky (51) were meta-analyses that primarily
included early generation vascular closure technologies,
primarily dating from the 1990s. More recent single site
propensity analyses by Arora (53) and Applegate (52)
demonstrated parity or superiority of vascular closure, albeit
at sites with very high volume experience. Abbreviations:
MC, manual compression; VCD, vascular closure device.
radial and ulnar arteries. The patient then makes a fist with his Access Technique
hands several times and then extends the palm demonstrating After placement of the oximeter on the ipsilateral thumb, the
blanching due to lack of adequate inflow. With release of the entire hand and forearm are prepped and care is taken to have
pressure over the ulnar artery, rapid resolution of the blanching a sterile environment under and around the forearm. Some
is a sign of an intact and widely patent palmar arch. By operators use a sterile sock placed over the hand to the wrist
convention, normal dual circulation is referred to as a “posi- through which a hole is cut at the radial access point. Although
tive” Allen test (56). The mean time for the vascular blush to some laboratories abduct the arm, this is usually not necessary,
normalize is relatively short, typically less than five seconds and placement of the patient’s straight arm against the thigh
(57). The Allen test has been criticized both for lack of specific- makes the distance to the radiation source similar to femoral
ity and sensitivity (58), but an analysis of the literature suggests access. It also eliminates a gap between the wrist and the cath
that it remains a useful guide over all (59). Most cath labs use a table that otherwise results in blood, saline, contrast and var-
modification of the Allen test, placing a pulse oximeter on the ious devices falling to the floor. Specialized drapes are avail-
thumb and monitoring oximetry (60), or oximetry and plethys- able, although many labs simply position the two “doughnut
mography combined (61), while compressing the radial artery. holes” so that one is over the right wrist and the other over the
The Allen test alone has somewhat reduced specificity: in over prepped right femoral, allowing easy conversion to femoral
1000 patients (57) a normal palmar flush in at least one hand access if necessary. The artery is most accessible without over-
could be detected within nine seconds in 93.6%. Using plethys- lying connective tissue if the wrist is not extended. Left arm
mography combined with oximetry, an oximeter placed on the access is preferred by some operators, in particular when the
thumb demonstrated preserved saturation of 90% or greater distance to the target vessel is long (such as for renal or lower
and an arterialized waveform immediately after radial occlu- extremity interventions), the left internal mammary artery is to
sion in 96.4% of the same patient cohort. An additional 2.1% be entered, or the right radial is inadequate. Catheter manip-
had no arterial waveform immediately after occlusion, but ulation into the coronary arteries is simpler via the left radial
within two minutes of continuous radial occlusion sufficient approach (64), and the success rates are equivalent to right
flow through the palmar arch occurred to achieve an arterial- radial catheterization (65), although slightly longer procedure
ized waveform in an additional 2.1%. Thus, only 1.5% of times have been reported (66). If the patient’s body habitus
patients had no discernable waveform at 120 seconds. The allows flexing the left hand over his or her abdomen, the
suitability of patients for radial procedures based on oximetry procedure can be performed from the right side of the table.
alone remains the subject of some debate (62), in part because Local anesthesia is injected intradermally approximately
oximetry may suggest adequate collateralization even when 2 cm proximal to the styloid process, using a 25-gauge or
flow is reduced by more than 90% (63). The use of plethysmog- smaller needle, and minimizing the subdermal volume of
raphy combined with oximetry has led to the Barbeau classifi- anesthetic to minimize distortion of the anatomy, inadvertent
cation, based on the data cited above (Fig. 8.11) (57). arterial entry, and potential induction of radial spasm. Radial
[13/7/010/12:55:54] [67–84]
has also been implemented for radial procedures (73). Given

that there is substantial discomfort reported by some patients
after intra-arterial heparin administration, intravenous admin-
istration should be considered. A randomized comparison of
500 patients failed to demonstrate any difference in radial
artery occlusion between direct intra-arterial versus intrave-
nous heparin application (74). Systemic enoxaprin also
appears to have similar rates of radial artery patency (75),
and transition from the initial heparin injected for arterial
patency to systemic bivalirudin for coronary intervention has
not been shown to impact bleeding complication rates (71); a
30-minute interval between heparin and bivalirudin adminis-
tration has been recommended (76). Most operators inject the
spasmolytic and anticoagulant cocktail in a solution diluted
with saline. Consideration should be given to filling the
remainder of the syringe with blood rather than saline (77),
injecting small amounts of the mixture, and diluting repeat-
edly by drawing back on the syringe, since blood is a more
effective buffer and may result in less patient discomfort.
Once access is obtained, and appropriate medication
Figure 8.11 The Barbeau classification. Four patterns of the Allen infused, consideration should be given to assessing systemic
test, as detected by plethysmography with temporary occlusion of pressure through the sheath sidearm. Pressures lower than the
the radial artery. (A) A normal “positive” Allen test, showing no contralateral arm cuff pressure will provide an early clue to
significant diminution of the waveform immediately after radial stenosis or occlusion of the ipsilateral radial or more proximal
occlusion. (B) The initial postocclusion waveform is present but arm circulation, or may imply spasm or hypoplasia of the arm
diminished, and returns to a full excursion within two minutes. (C) vessels. Higher pressures may suggest a subclavian or axillary
No waveform is detected immediately after occlusion, but some stenosis in the contralateral arm.
phasic waveform is seen within two minutes. (D) There is no flow Three distinct anatomic segments need to be traversed:
detected by plethysmorgraphy in class D even after two minutes of from the radial to the axillary artery, the axillary to the
arterial occlusion. Oximetry is expected to demonstrate saturation of
innominate, and from the transverse arch to the coronary
90% or greater with a waveform present. Note that patients with a
type C response, presumably after recruitment of collaterals, will cusps. The first segment requires dealing with vessel caliber,
frequently exhibit a type B response if the test is repeated early. tortuosity and anatomical variance issues; the second with tor-
Source: From Ref. 57. tuosity, sidebranches and occasionally occlusion; the third with
steering away from the descending aorta and into the cusps
without excessive manipulation. Catheter advancement through
the sheath should be with the aid of a guidewire; the type of
access uses a micropuncture technique in most laboratories, guidewire is the source of considerable disagreement among
typically a 4 cm long 21-gauge needle (rather than the typical operators. The choices vary from standard 0.035@ straight or J-
7-cm needle used for femoral access). Bleed back through the tipped stainless steel guidewires, to hydrophilic-coated angled
needle, typically a trickle of blood, is followed by introduction tip wires, to 0.014@ coronary wires (78). Regardless of wire type
of a micropuncture wire, ranging from 0.018@ to 0.025@ diam- chosen, advancement should only take place if there is no resis-
eter. On occasion, tortuosity of the radial artery does not allow tance, and ideally should be guided by fluoroscopy, particularly
conventional micropuncture wire access, and a 0.014@ coronary if a hydrophilic or coronary guidewire is used. Advancement into
guidewire will successfully traverse the forearm. A number of radial, ulnar, or brachial sidebranches, particularly against resis-
sheaths have been utilized for radial access, and because of the tance, but even without, can result in perforation and compart-
smaller caliber of the radial artery as well as the potential for ment syndromes as discussed subsequently.
spasm, most operators prefer hydrophilic-coated sheaths (67). A variety of anatomical variations may require extensive
These are 4F to 8F in diameter (typically 5–6F), with step-up manipulation, use of smaller, sometimes coronary wires, and
dilators, and range in length from 4 to 30 cm, the latter used by occasionally a buddy wire system (79); regardless, considerable
some operators to reduce the potential for spasm in the forearm care is required to avoid arterial perforation or dissection, or
caused by torquing of catheters as they traverse the radial and catheter entrapment. The most common variations are high
brachial vessels. radial bifurcation (usually not an impediment to catheteriza-
After sheath introduction, most operators inject a “cock- tion), radioulnar loops, severe tortuosity, and hypoplasia, and
tail” of one or more medications. These typically include occur in up to 10% to 15% of patients (80), although the
vasodilators to minimize spasm and anticoagulants to prevent majority of these can be traversed safely (81). In a small
thrombosis. Although the evidence base is incomplete, the percentage of cases, catheter passage to the shoulder is impos-
most studied have been intra-arterial nitroglycerin and sible or best avoided; a significant learning curve exists (11)
verapamil (68,69), with doses in the range of 100 to 250 mg (Fig. 8.12). Once the guidewire is in the axillary and subclavian
for the former, and 1 to 5 mg for the latter. Heparin has been arteries, the operator needs to avoid major head and neck
given at doses typically in the range of 1000 to 5000 units vessels. Entry into the vertebral and carotid circulations is
(70,71), with some evidence that patency is related to heparin particular common, especially with hydrophilic wires. Tortuos-
dosage (72). Weight-based dosing, an important factor in ity of the right subclavian and innominate is particularly com-
decreasing bleeding complications by the femoral route (35), mon in the elderly, and care should be taken to minimize
[13/7/010/12:55:54] [67–84]
Figure 8.12 (A) Normal brachial arterial distribution anatomy showing bifurcation below the elbow and a prominent interosseus vessel. The
injection was prior to instillation of a verapamil and heparin “cocktail.” (B) Radial artery loop with recurrent radial circulation. A radial artery
loop (full 3608 turn of the artery distal to the brachial bifurcation) occurs in approximately 2% of patients, and represents the most common
cause for access failure for experienced radial catheterizers. Source: Figure 8.12 (B) from Ref. 80 with courtesy of BMJ Publishing Group Ltd.
guidewire manipulation to avoid dislodging plaque. Once the vein in most cases, a technique that is particularly appealing in
subclavian has been fully negotiated, there is a tendency to fully anticoagulated patients where there is contraindication to
enter the descending aorta. Moving to a left anterior oblique discontinuation of anticoagulation (83).
view helps identify catheter passage into the ascending aorta.
Because traversing the arm circulation requires more
effort than the typical iliac circulation, most operators attempt Closure
to use a single catheter for diagnostic coronary angiography. Because of the relatively long compression times required for
Although a number of catheter curves are made for this pur- hemostasis, various mechanical compression devices are used
pose, an ideal catheter shape has been elusive. Exchange length (84–86). The compression time is variable, and ranges from as
guidewires should be used to switch catheters when this is little as 30 minutes (or less) to as long as four hours, with the
necessary; prolonged dwell time of guidewires in the subcla- longer intervals typically recommended for fully anticoagu-
vian and innominate circulation is a risk for cranial emboliza- lated patients. Hand ischemia occurs if mechanical compres-
tion, and multiple catheter passages through the arm sion devices occlude both the radial and ulnar arteries.
circulation tend to provoke spasm. Manipulation of catheters Excessive pressure can result in vascular soft tissue and nerve
in the coronary cusps requires different technique than femoral injury. Of particular importance is evidence (see below) that
access, and as a rule, left coronary catheter curves need to be a compression of the artery with only partial occlusion of blood
half size smaller when approaching from the right arm. flow is important for maintaining long-term patency.
Because of the smaller caliber of radial arteries [typically
in the 3–4 mm range, compared with 6–8 mm for the average
femoral (20,82)], diagnostic catheterization is typically per- Complications
formed with 5F catheters, although some operators use 4F. Although one of the main advantages of the radial approach is
These require higher pressure to obtain adequate contrast flow the lower complication rate, some adverse events are associated
into the coronary arteries for optimal visualization. Catheter with this approach. These include occlusion of the radial artery,
manipulation sometimes requires guidewires to be left in place spasm, bleeding, compartment syndrome, hand ischemia, dis-
until the coronary is intubated to prevent catheter kinking and section and other trauma to the head and neck vessels, embo-
to facilitate torque control. lization, and a myriad of rare complications including
An occasional variant to radial access is ulnar access; the arteriovenous fistula, major hematoma, transfusion, stroke
ulnar artery is typically smaller and more likely to be associated and pseudoaneurysm.
with spasm (82). Finally, both right and left heart catheter- Occlusion of the radial artery as a consequence of cath-
ization are done from the arm by an increasing number of eter placement has been reported to range from 1% to 38%, the
operators, using the right radial artery and right antecubital higher range being associated with prolonged dwell times of
[13/7/010/12:55:54] [67–84]
radial artery catheters used for monitoring; the generally Compartment syndrome is a rare but dreaded complica-
accepted value after cardiac catheterization is in the range of tion of radial access (95), occurring in substantially less than 1%
3% to 5% (8,87). There is evidence that up to half of patients of patients. Any apparent perforation, particularly in the fore-
with occlusions recanalize in the month following the catheter- arm, should be a cause for concern, and if possible should
ization. Most occlusions are not clinically apparent, and thus result in reversal of anticoagulation. A suggested algorithm has
are not counted as complications in most series. Reduction of been to promptly apply pressure at a level 10 to 15 mmHg
flow after sheath removal appears related to a number of below systemic systolic so as to allow distal flow, and maintain
factors, including the ratio between sheath size and arterial compression for 15-minute periods (95). Vigilance, a high index
diameter (88), extent of anticoagulation, degree of transient of suspicion, and prompt intervention are required to prevent
occlusion of the radial artery, and number of arterial accesses. permanent injury.
There is some evidence that prolonged interruption of radial A persistent concern has been the need to preserve the
artery flow for hemostasis exacerbates the problem of chronic radial artery for future catheterization and potentially for
radial artery occlusion (89). In a study using plethysmography harvesting for coronary bypass grafting. The artery does
(73), patients were randomized to two hours of total occlusion develop intimal hyperplasia and decreased vasomotor tone
versus two hours of compression with preserved radial artery after use for catheter access (96), with inconsistent findings
flow; the latter was associated with a decrease in long-term regarding the apparent long-term affect of catheterization on
radial occlusion by an odds ratio of nearly 4 to 1. Similarly, a radial artery physiologic function (97,98).
randomized trial compared inflation of a compression device
over the radial artery to mean arterial pressure (102.5 mmHg)
compared with injection of a fixed volume of air (device pres- Radial Compared with Femoral Access
sure 185 mmHg). The study was stopped prematurely because A large evidence base, including randomized comparisons, has
the radial artery occlusion rate was only 1.1% with the former, addressed the issue of radial versus femoral access (11,99). A
compared with 12% for the latter (90). meta-analysis by Agostini (11) of 12 randomized trials enrolling
Spasm can cause pain, limit torque control, and on occa- over 3000 patients demonstrated superiority of the radial
sion require conversion to femoral access. It can result in approach in access-related complication rates (0.3% vs. 2.8%),
catheter trapping, with occasional cases of radial artery evul- but higher failure rates (7.2% vs. 2.4%). The latter is due to
sion being recorded (Fig. 8.13). Hydrophilic sheaths have failed cannulation of the radial artery, failure to gain access to
resulted in a reduction of spasm, with less resistance to with- the ascending aorta through complex arm or shoulder anat-
drawal (91). Lower force required to withdraw the sheath as omy, or failure to intubate the coronary arteries.
well as less pain have also been demonstrated in a randomized The largest nonrandomized comparison, using the ACC-
trial (92). Some coatings, although they reduce the coefficient of NCDR database, compared nearly 600,000 coronary interven-
friction, have potential associated toxicity (93,94), in particular tions via the femoral approach with close to 8000 via the radial
sterile abscess formation, likely as an allergic response. approach and found evidence of a greater than 2:1 odds ratio
for hemorrhagic complications with femoral access. Moreover,
this relationship was consistent, statistically or with a strong
trend, across age groups and in settings such as ST-elevation
myocardial infarction. The benefits of radial access in acute
ST-elevation myocardial infarction have been confirmed in
several studies (100,101), with a modest percentage of patients
crossing over to femoral access to prevent significant delay in
door-to-balloon time when radial access was not immediate.
The radial approach has particular appeal in rescue angioplasty
after failed thrombolytic therapy (102). In patients over age 80,
the advantages of the radial approach remain striking (103).
The learning curve has played a substantial role, and the
more recent trials included in the meta-analyses did not show a
difference in success rates between femoral and radial access
(11). In a review of radial access success rates, the failure rate
was greatest in an operator’s first 20 cases (14%), but decreased
to 4% when experience exceeded 100 cases; concomitant fluo-
roscopy time decreased from a mean of 9.6 to 6.8 minutes (104).
Figure 8.14 shows the influence of the learning curve on
fluoroscopy times and success rates in the hands of a single
operator.
There is evidence of strong patient preference for the
radial technique (87,106), in large part because it avoids the
immobilization associated with femoral closure, as well as the
Figure 8.13 Catheter entrapment and simultaneous hand ische- discomfort associated with manual compression of the femoral
mia. Contrast injection demonstrates no flow around the catheter tip artery (66) and with many VCDs. Costs associated with radial
(white arrow). The patient’s hand became ischemic [note severe catheterization are lower, in part because of earlier ambulation,
stenosis in the ulnar artery (black arrow)]. The catheter was lower complication rates (106,107), and the lower cost of radial
extracted surgically. compression devices than most femoral VCDs (108). The
lower cost of the radial approach holds true for diagnostic
[13/7/010/12:55:54] [67–84]
Figure 8.14 Relationship between operator experience, procedure time and technical failure for radial artery access. Source: From Ref. 105.
catheterization as well, with or without VCD use (109). Axillary puncture, once relatively common, has become
Although procedure time is general slightly longer with radial rare, primarily because of the risk of brachial plexus injury.
access, this has not been consistent, and is likely related to Direct translumbar puncture into the aorta, although described
learning curve issues as well. An important issue, radiation in diagnostic catheterization and even stent placement (119),
exposure, has been shown to be greater with radial than has rarely been performed. A procedure making a modest
femoral access (66), although efficacy of shielding and operator return is direct transapical puncture (120,121), once performed
experience may modify or potentially nullify the difference primarily for hemodynamic evaluation in patients who had
(110). tilting disk type mechanical valves in both mitral and aortic
positions, and still of value when noninvasive data are incon-
clusive. However, it is the performance of various structural
OTHER ACCESS ROUTES heart disease interventions, in particular percutaneous closure
The era of brachial cutdown for vascular access dates to the
of prosthetic paravalvular leaks (122), that is providing some
earliest catheterization era, and was integral to the work of
additional utilization of this procedure. It is typically per-
pioneers such as Sones (111). Brachial cutdown is now seldom
formed with ultrasound guidance, and direct chest wall com-
performed, and when brachial access is selected, is typically via
pression on catheter removal; the complication rate (all in
a percutaneous route. Percutaneous access for cardiac catheter-
patients who had undergone prior thoracotomy) has been
ization was implemented in the 1980s (112,113), including the
reported to range from 0.5% to 3% (123).
use of sheaths (114). The brachial artery remains a viable
alternative when larger catheters are required, the palmar
circulation is inadequate for radial access, or catheter length VENOUS ACCESS
is inadequate to allow use of the radial artery. Important The primary techniques for venous access in the cardiac
anatomic issues include the location of the median nerve catheterization laboratory are via the femoral and internal
immediately adjacent to the artery, increasing the risk associ- jugular veins. Venous access by the brachial route [with the
ated with hematoma and secondary nerve compression. Access exception of some laboratories that routinely do combined right
is typically several centimeters proximal to the crease of the radial and right basilic catheterization (83)] by direct puncture
elbow; more proximal puncture results in deeper arterial entry, or cutdown, the external jugular, and the subclavian are
with more muscle and connective tissue between the skin uncommon, and rarer still are axillary, cephalic, popliteal,
surface and the humerus, resulting in more difficult compres- percutaneous transhepatic and other ports of venous entry,
sion and higher hematoma and pseudoaneurysm rates. although some are more commonly accessed by interventional
Unlike catheterization via the radial or ulnar arteries with radiologists. Right heart catheterization, once a ubiquitous part
a positive Allen test, hand ischemia is an important potential of all heart catheterization procedures, was used as a marker of
complication of brachial access (115), albeit occurring less inappropriate procedure selection in the 1980s, and is now part
frequently than early in the utilization of this technique of a minority of heart catheterizations. Nevertheless, in addi-
(114,116). Multiple retrospective reviews have documented a tion to diagnostic right heart catheterization, endomyocardial
higher complication rate with brachial than with femoral access biopsy, and occasional transvenous pacemaker placement,
(117), exacerbated by learning curve issues in operators who venous access remains important for a wide range of electro-
typically catheterize primarily via the femoral route (118). A physiology procedures and structural heart disease interven-
900 patient randomized trial comparing radial, brachial, and tions.
femoral access demonstrated success rates of 93.0%, 95.7%, and Femoral venous access remain the most common route in
99.7%, respectively, and major access site related complications in the cardiac catheterization laboratory because of the conve-
0%, 2.3%, and 2.0%, respectively. Thus, the brachial approach has nience of using the same site as arterial puncture (in case of
the highest complication rate, and an intermediate success rate simultaneous left and right heart catheterization), the large size
when compared with radial and femoral access (8). and distensibility of the femoral vein, and relative safety and
[13/7/010/12:55:54] [67–84]
simplicity. Contraindications include thrombosis of the ipsi- to facilitate hemostasis (130). The femoral vein is relatively
lateral iliac vein, the inferior vena cava, or infection at the fragile, and care needs to be taken to avoid excessive traction
inguinal access site. The presence of a vena caval filter does not on the vein during suture closure to prevent severing the vessel,
preclude the passage of catheters or sheaths, although some a potentially serious adverse event, since virtually all venous
operators will inject contrast in this setting to assure absence of efflux from the leg is through the common femoral vein. A novel
obvious thrombus that might result in embolization. Femoral “figure-of-eight” alternative has been described that places a
venous access uses similar landmarks as those discussed for the deep suture around a large femoral vein catheter (without
femoral artery, with the vein located one-half to one cm medial entering or passing deep to the femoral vein itself), capturing
to the artery. We use fluoroscopy, but the evidence base is most substantial tissue and compressing the puncture site when the
compelling for ultrasound-guided access (124). Although some sheath is pulled and tension is applied to the suture (131).
operators choose a lower puncture site for venous than for
arterial access, we believe the same considerations apply,
namely, avoidance of low puncture because of increased risk CONCLUSIONS
of arteriovenous fistula and more difficult compression, and Vascular access has evolved over a half century with the devel-
avoidance of high puncture because of the risk of RPH. Val- opment of a substantial evidence base, in particular in the past
salva maneuver and pressure on the patient’s abdomen can 15 years. Use of adjunctive technologies for access, in particular
distend the femoral vein and make access easier; hydration, ultrasound and fluoroscopy, micropuncture, and modifications
especially for a patient who has taken nothing by mouth for a in anticoagulation and antiplatelet therapy has combined to
prolonged period, can increase the diameter of the vein as well. improve access success and reduce overall complication rates.
Most operators use 18-gauge needle access for the fem- Awareness of anatomy, and routine use of femoral angiography
oral vein, with similar methodology as for the modified Sel- can avoid or potentially help address a number of complica-
dinger technique, using a hollow needle and puncture of the tions. A transition from primarily femoral to primarily radial
anterior wall of the vein. Because of low pressure in the venous access is taking place worldwide, although the femoral route
system, continuous aspiration through the hollow needle helps remains dominant in the United States. VCDs, without a com-
identify vessel entry. We use a micropuncture technique in pelling evidence base to justify routine use, improve patient
venous access as well. Inadvertent entry into the femoral vein is comfort and convenience after femoral access. The expanding
relatively common during planned arterial access [occurring in role of structural heart disease interventions will place further
up to 30% of cases in one study (27)]; in this event, we leave a emphasis on safe access for large diameter catheters, and should
wire in place in the vein during arterial access as a fluoroscopic lead to development of additional novel techniques for percuta-
marker to guide entry into the artery. neous closure.
Internal jugular access avoids the problems associated
with the femoral triangle, facilitates early ambulation, provides
a superior angle of entry into the right ventricle for right heart REFERENCES
catheterization, and provides for superior control during right 1. Seldinger SI. Catheter replacement of the needle in percutaneous
ventricular endomyocardial biopsy. Ultrasound guidance sub- arteriography; a new technique. Acta Radiol 1953; 39:368–376.
stantially improves the time to access, success rate, avoidance 2. Sones FM Jr., Shirey EK. Cine coronary arteriography. Mod
Concepts Cardiovasc Dis 1962; 31:735–738.
of inadvertent entry into adjacent structures, and lowers the
3. Dotter CT, Judkins MP. Transluminal treatment of arterioscle-
overall complication profile of the procedure (124). The internal
rotic obstruction. Description of a new technic and a preliminary
jugular vein courses in close proximity to the carotid artery and report of its application. Circulation 1964; 30:654–670.
inadvertent carotid intubation is relatively common (in the 4. Judkins MP. Selective coronary arteriography. I. A percutaneous
range of 3–4%) in the absence of two dimensional ultrasound transfemoral technic. Radiology 1967; 89:815–824.
guidance (125). Carotid puncture, especially with a large nee- 5. Campeau L. Percutaneous radial artery approach for coronary
dle, has a wide range of associated risk, including cerebrovas- angiography. Cathet Cardiovasc Diagn 1989; 16:3–7.
cular events. Associated hematoma can be difficult to control 6. Kiemeneij F, Laarman GJ. Percutaneous transradial artery
with rare cases of tracheal compression. An additional impor- approach for coronary stent implantation. Cathet Cardiovasc
tant complication is pneumothorax, associated with inadvertent Diagn 1993; 30:173–178.
7. Pygott F, Hutton CF. Vertebral arteriography by percutaneous
puncture of the lung, sometimes associated with a high pul-
brachial artery catheterisation. Br J Radiol 1959; 32:114–119.
monary bleb in patients with chronic lung disease. While
8. Kiemeneij F, Laarman GJ, Odekerken D, et al. A randomized
subclavian access is relatively safe, the general consensus is comparison of percutaneous transluminal coronary angioplasty
that internal jugular access is safer, although the evidence base by the radial, brachial and femoral approaches: the access study.
is not compelling (126). J Am Coll Cardiol 1997; 29:1269–1275.
Venous closure is almost always by compression, 9. Barry WH, Levin DC, Green LH, et al. Left heart catheterization
although in various settings adjunctive noninvasive closure and angiography via the percutaneous femoral approach using
methods, such as topical patches, have been utilized (127). an arterial sheath. Cathet Cardiovasc Diagn 1979; 5:401–409.
Care should be taken to avoid excessive compression time, 10. Eggebrecht H, Oldenburg O, Dirsch O, et al. Potential emboliza-
since venous thrombosis remains a potential complication tion by atherosclerotic debris dislodged from aortic wall during
cardiac catheterization: histological and clinical findings in 7,621
(128), likely related to duration of compression with total occlu-
patients. Catheter Cardiovasc Interv 2000; 49:389–394.
sion of the femoral vein. Various VCDs have been used for
11. Agostoni P, Biondi-Zoccai GG, de Benedictis ML, et al. Radial
femoral venous closure (offlabel) (129), but the risk-to-benefit versus femoral approach for percutaneous coronary diagnostic
ratio is not established, and given the low pressure in the venous and interventional procedures; Systematic overview and meta-
circulation there is no compelling case for routine use. For some analysis of randomized trials. J Am Coll Cardiol 2004; 44:349–356.
structural heart disease interventions, where large sheaths are 12. Turi ZG. Overview of vascular closure. Endovasc Today 2009;
required (14F or greater), a “preclosure” technique can be used 9:24–32.
[13/7/010/12:55:54] [67–84]
13. Grier D, Hartnell G. Percutaneous femoral artery puncture: 33. Sohail MR, Khan AH, Holmes DR Jr., et al. Infectious complica-
practice and anatomy. Br J Radiol 1990; 63:602–604. tions of percutaneous vascular closure devices. Mayo Clin Proc
14. Garrett PD, Eckart RE, Bauch TD, et al. Fluoroscopic localization 2005; 80:1011–1015.
of the femoral head as a landmark for common femoral artery 34. Ndrepepa G, Berger PB, Mehilli J, et al. Periprocedural bleeding
cannulation. Catheter Cardiovasc Interv 2005; 65:205–207. and 1-year outcome after percutaneous coronary interventions:
15. Davidson CJ, Bonow RO. Cardiac catheterization. In: Libby P, appropriateness of including bleeding as a component of a
Bonow RO, Mann DL, et al., eds. Braunwald’s Heart Disease. quadruple end point. J Am Coll Cardiol 2008; 51:690–697.
Philadelphia: Saunders Elsevier, 2008:439–463. 35. Blankenship JC, Balog C, Sapp SK, et al. Reduction in vascular
16. Kim D, Orron DE, Skillman JJ, et al. Role of superficial femoral access site bleeding in sequential abciximab coronary interven-
artery puncture in the development of pseudoaneurysm and tion trials. Catheter Cardiovasc Interv 2002; 57:476–483.
arteriovenous fistula complicating percutaneous transfemoral 36. Use of a monoclonal antibody directed against the platelet
cardiac catheterization [see comments]. Cathet Cardiovasc glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty.
Diagn 1992; 25:91–97. The EPIC Investigation [see comments]. N Engl J Med 1994;
17. Altin RS, Flicker S, Naidech HJ. Pseudoaneurysm and arterio- 330:956–961.
venous fistula after femoral artery catheterization: association 37. Randomised placebo-controlled and balloon-angioplasty-con-
with low femoral punctures. AJR Am J Roentgenol 1989; 152: trolled trial to assess safety of coronary stenting with use of
629–631. platelet glycoprotein-IIb/IIIa blockade. Lancet 1998; 352:87–92.
18. Ellis SG, Bhatt D, Kapadia S, et al. Correlates and outcomes of 38. Dauerman HL, Applegate RJ, Cohen DJ. Vascular closure devi-
retroperitoneal hemorrhage complicating percutaneous coronary ces: the second decade. J Am Coll Cardiol 2007; 50:1617–1626.
intervention. Catheter Cardiovasc Interv 2006; 67:541–545. 39. Cura FA, Kapadia SR, L’Allier PL, et al. Safety of femoral closure
19. Turi ZG. Overview of vascular closure. Endovasc Today 2008; devices after percutaneous coronary interventions in the era of
7:28–37. glycoprotein IIb/IIIa platelet blockade. Am J Cardiol 2000;
20. Schnyder G, Sawhney N, Whisenant B, et al. Common femoral 86:780–782.
artery anatomy is influenced by demographics and comorbidity: 40. Tiroch KA, Arora N, Matheny ME, et al. Risk predictors of
implications for cardiac and peripheral invasive studies. Catheter retroperitoneal hemorrhage following percutaneous coronary
Cardiovasc Interv 2001; 53:289–295. intervention. Am J Cardiol 2008; 102:1473–1476.
21. Turi ZG, McEniry BJ, Turi MN. A quantitative angiographic 41. Silva JA, Stant J, Ramee SR. Endovascular treatment of a massive
evaluation of the target zone for vascular access. Am J Cardiol retroperitoneal bleeding: successful balloon-catheter delivery of
2007; 100(suppl 1):S229 (abstr). intra-arterial thrombin. Catheter Cardiovasc Interv 2005; 64:
22. Dotter CT, Rosch J, Robinson M. Fluoroscopic guidance in 218–222.
femoral artery puncture. Radiology 1978; 127:266–267. 42. Olsen DM, Rodriguez JA, Vranic M, et al. A prospective study of
23. Fitts J, Ver LP, Hofmaster P, et al. Fluoroscopy-guided femoral ultrasound scan-guided thrombin injection of femoral pseudoa-
artery puncture reduces the risk of PCI-related vascular compli- neurysm: a trend toward minimal medication. J Vasc Surg 2002;
cations. J Interv Cardiol 2008; 21:273–278. 36:779–782.
24. Abu-Fadel MS, Sparling JM, Zacharias SJ, et al. Fluoroscopy vs. 43. Mohler ER III, Mitchell ME, Carpenter JP, et al. Therapeutic
traditional guided femoral arterial access and the use of closure thrombin injection of pseudoaneurysms: a multicenter experi-
devices: a randomized controlled trial. Catheter Cardiovasc ence. Vasc Med 2001; 6:241–244.
Interv 2009; 74:533–539. 44. Kalapatapu VR, Ali AT, Masroor F, et al. Techniques for manag-
25. Turi ZG. Fluoroscopy guided vascular access: asking the right ing complications of arterial closure devices. Vasc Endovascular
question, but getting the wrong answer? Catheter Cardiovasc Surg 2006; 40:399–408.
Interv 2009; 74:540–542. 45. Tavris DR, Dey S, Albrecht-Gallauresi B, et al. Risk of local
26. Dudeck O, Teichgraeber U, Podrabsky P, et al. A randomized adverse events following cardiac catheterization by hemostasis
trial assessing the value of ultrasound-guided puncture of the device use—phase II. J Invasive Cardiol 2005; 17:644–650.
femoral artery for interventional investigations. Int J Cardiovasc 46. Balzer JO, Scheinert D, Diebold T, et al. Postinterventional trans-
Imaging 2004; 20:363–368. cutaneous suture of femoral artery access sites in patients with
27. Seto AH, Abu-Fadel MS, Sparling JM, et al. Real-time ultrasound peripheral arterial occlusive disease: a study of 930 patients.
facilitates femoral aterial access and reduces vascular complica- Catheter Cardiovasc Interv 2001; 53:174–181.
tions: the Femoral Arterial Access With Ultrasound (FAUST) 47. Warren BS, Warren SG, Miller SD. Predictors of complications
trial. Presented at Transcatheter Cardiovascular Therapeutics, and learning curve using the Angio-Seal closure device following
September 25, 2009; San Francisco, CA. interventional and diagnostic catheterization. Catheter Cardio-
28. Stewart JH, Cole GW, Klein JA. Neutralized lidocaine with vasc Interv 1999; 48:162–166.
epinephrine for local anesthesia. J Dermatol Surg Oncol 1989; 48. Lee WA, Brown MP, Nelson PR, et al. Total percutaneous access
15:1081–1083. for endovascular aortic aneurysm repair (“Preclose” technique).
29. Farouque HM, Tremmel JA, Raissi SF, et al. Risk factors for the J Vasc Surg 2007; 45:1095–1101.
development of retroperitoneal hematoma after percutaneous 49. Resnic FS, Arora N, Matheny M, et al. A cost-minimization
coronary intervention in the era of glycoprotein IIb/IIIa inhib- analysis of the angio-seal vascular closure device following
itors and vascular closure devices. J Am Coll Cardiol 2005; percutaneous coronary intervention. Am J Cardiol 2007; 99:
45:363–368. 766–770.
30. Chandrasekar B, Doucet S, Bilodeau L, et al. Complications of 50. Koreny M, Riedmuller E, Nikfardjam M, et al. Arterial puncture
cardiac catheterization in the current era: a single-center experi- closing devices compared with standard manual compression
ence. Catheter Cardiovasc Interv 2001; 52:289–295. after cardiac catheterization: systematic review and meta-analy-
31. Sherev DA, Shaw RE, Brent BN. Angiographic predictors of sis. JAMA 2004; 291:350–357.
femoral access site complications: implication for planned per- 51. Nikolsky E, Mehran R, Halkin A, et al. Vascular complications
cutaneous coronary intervention. Catheter Cardiovasc Interv associated with arteriotomy closure devices in patients undergo-
2005; 65:196–202. ing percutaneous coronary procedures: a meta-analysis. J Am
32. Lenartova M, Tak T. Iatrogenic pseudoaneurysm of femoral Coll Cardiol 2004; 44:1200–1209.
artery: case report and literature review. Clin Med Res 2003; 52. Applegate RJ, Sacrinty MT, Kutcher MA, et al. Propensity score
1:243–247. analysis of vascular complications after diagnostic cardiac
[13/7/010/12:55:54] [67–84]
catheterization and percutaneous coronary intervention 1998– 75. Feray H, Izgi C, Cetiner D, et al. Effectiveness of enoxaparin for
2003. Catheter Cardiovasc Interv 2006; 67:556–562. prevention of radial artery occlusion after transradial cardiac
53. Arora N, Matheny ME, Sepke C, et al. A propensity analysis of catheterization. J Thromb Thrombolysis 2009. e-pub May 29.
the risk of vascular complications after cardiac catheterization 76. White HD, Chew DP, Hoekstra JW, et al. Safety and efficacy of
procedures with the use of vascular closure devices. Am Heart J switching from either unfractionated heparin or enoxaparin to
2007; 153:606–611. bivalirudin in patients with non-ST-segment elevation acute
54. Rao SV, Ou FS, Wang TY, et al. Trends in the prevalence and coronary syndromes managed with an invasive strategy: results
outcomes of radial and femoral approaches to percutaneous from the ACUITY (Acute Catheterization and Urgent Interven-
coronary intervention; A report from the National Cardiovascular tion Triage strategy) trial. J Am Coll Cardiol 2008; 51:1734–1741.
Data Registry. JACC Cardiovasc Interv 2008; 1:379–386. 77. Patel T, Shah S, Ranjan A. Puncture technique. In: Patel T, ed.
55. Allen EV. Thromboangiitis obliterans; methods of diagnosis of Patel’s Atlas of Transradial Intervention. Seattle: Sea Script Com-
chronic occlusive arterial lesions distal to the wrist with illustra- pany, 2007:17.
tive cases. Am J Med Sci 1929; 178:237–244. 78. Barbeau GR. Radial loop and extreme vessel tortuosity in the
56. Peters KR, Chapin JW. Allen’s test–positive or negative? Anes- transradial approach: advantage of hydrophilic-coated guide-
thesiology 1980; 53:85. wires and catheters. Catheter Cardiovasc Interv 2003; 59:442–450.
57. Barbeau GR, Arsenault F, Dugas L, et al. Evaluation of the 79. Dana A, Barbeau GR. The use of multiple “buddies” during
ulnopalmar arterial arches with pulse oximetry and plethysmog- transradial angioplasty in a complex calcified coronary tree.
raphy: comparison with the Allen’s test in 1010 patients. Am Catheter Cardiovasc Interv 2006; 67:396–399.
Heart J 2004; 147:489–493. 80. Lo TS, Nolan J, Fountzopoulos E, et al. Radial artery anomaly
58. Barone JE, Madlinger RV. Should an Allen test be performed and its influence on transradial coronary procedural outcome.
before radial artery cannulation? J Trauma 2006; 61:468–470. Heart 2009; 95:410–415.
59. Ronald A, Patel A, Dunning J. Is the Allen’s test adequate to safely 81. Yokoyama N, Takeshita S, Ochiai M, et al. Anatomic variations of
confirm that a radial artery may be harvested for coronary arterial the radial artery in patients undergoing transradial coronary
bypass grafting? Interact Cardiovasc Thorac Surg 2005; 4:332–340. intervention. Catheter Cardiovasc Interv 2000; 49:357–362.
60. Hovagim AR, Katz RI, Poppers PJ. Pulse oximetry for evaluation 82. Vassilev D, Smilkova D, Gil R. Ulnar artery as access site for
of radial and ulnar arterial blood flow. J Cardiothorac Anesth cardiac catheterization: anatomical considerations. J Interv Car-
1989; 3:27–30. diol 2008; 21:56–60.
61. Matsuki A. A modified Allen’s test using a pulse oxymeter. 83. Lo TS, Buch AN, Hall IR, et al. Percutaneous left and right heart
Anaesth Intensive Care 1988; 16:126–127. catheterization in fully anticoagulated patients utilizing the
62. Levinsohn DG, Gordon L, Sessler DI. The Allen’s test: analysis of radial artery and forearm vein: a two-center experience. J Interv
four methods. J Hand Surg Am 1991; 16:279–282. Cardiol 2006; 19:258–263.
63. Lawson D, Norley I, Korbon G, et al. Blood flow limits and pulse 84. Chatelain P, Arceo A, Rombaut E, et al. New device for com-
oximeter signal detection. Anesthesiology 1987; 67:599–603. pression of the radial artery after diagnostic and interventional
64. Kawashima O, Endoh N, Terashima M, et al. Effectiveness of cardiac procedures. Cathet Cardiovasc Diagn 1997; 40:297–300.
right or left radial approach for coronary angiography. Catheter 85. Kiemeneij F. Radial hemostasis devices: alternatives for a simple pile
Cardiovasc Interv 2004; 61:333–337. of gauze and elastic plasters? J Invasive Cardiol 2000; 12:623–624.
65. Santas E, Bodi V, Sanchis J, et al. The left radial approach in daily 86. Pancholy SB. Impact of two different hemostatic devices on
practice. A randomized study comparing femoral and right and radial artery outcomes after transradial catheterization. J Inva-
left radial approaches. Rev Esp Cardiol 2009; 62:482–490. sive Cardiol 2009; 21:101–104.
66. Louvard Y, Lefevre T, Allain A, et al. Coronary angiography 87. Stella PR, Kiemeneij F, Laarman GJ, et al. Incidence and outcome
through the radial or the femoral approach: the CARAFE study. of radial artery occlusion following transradial artery coronary
Catheter Cardiovasc Interv 2001; 52:181–187. angioplasty. Cathet Cardiovasc Diagn 1997; 40:156–158.
67. Dery JP, Simard S, Barbeau GR. Reduction of discomfort at 88. Saito S, Ikei H, Hosokawa G, et al. Influence of the ratio between
sheath removal during transradial coronary procedures with radial artery inner diameter and sheath outer diameter on radial
the use of a hydrophilic-coated sheath. Catheter Cardiovasc artery flow after transradial coronary intervention. Catheter
Interv 2001; 54:289–294. Cardiovasc Interv 1999; 46:173–178.
68. Abe S, Meguro T, Endoh N, et al. Response of the radial artery to 89. Sanmartin M, Gomez M, Rumoroso JR, et al. Interruption of
three vasodilatory agents. Catheter Cardiovasc Interv 2000; blood flow during compression and radial artery occlusion after
49:253–256. transradial catheterization. Catheter Cardiovasc Interv 2007;
69. Kiemeneij F, Vajifdar BU, Eccleshall SC, et al. Evaluation of a 70:185–189.
spasmolytic cocktail to prevent radial artery spasm during cor- 90. Cubero JM, Lombardo J, Pedrosa C, et al. Radial compression
onary procedures. Catheter Cardiovasc Interv 2003; 58:281–284. guided by mean artery pressure versus standard compression
70. Chen CW, Lin CL, Lin TK, et al. A simple and effective regimen with a pneumatic device (RACOMAP). Catheter Cardiovasc
for prevention of radial artery spasm during coronary catheter- Interv 2009; 73:467–472.
ization. Cardiology 2006; 105:43–47. 91. Saito S, Tanaka S, Hiroe Y, et al. Usefulness of hydrophilic
71. Venkatesh K, Mann T. Transitioning from heparin to bivalirudin coating on arterial sheath introducer in transradial coronary
in patients undergoing ad hoc transradial interventional proce- intervention. Catheter Cardiovasc Interv 2002; 56:328–332.
dures: a pilot study. J Invasive Cardiol 2006; 18:120–124. 92. Kiemeneij F, Fraser D, Slagboom T, et al. Hydrophilic coating
72. Spaulding C, Lefevre T, Funck F, et al. Left radial approach for aids radial sheath withdrawal and reduces patient discomfort
coronary angiography: results of a prospective study. Cathet following transradial coronary intervention: a randomized dou-
Cardiovasc Diagn 1996; 39:365–370. ble-blind comparison of coated and uncoated sheaths. Catheter
73. Pancholy S, Coppola J, Patel T, et al. Prevention of radial artery Cardiovasc Interv 2003; 59:161–164.
occlusion-patent hemostasis evaluation trial (PROPHET study): a 93. Cogliano MA, Tolerico PH. Nonhealing wound resulting from a
randomized comparison of traditional versus patency docu- foreign body to a radial arterial sheath and sterile inflammation
mented hemostasis after transradial catheterization. Catheter associated with transradial catheterization and hydrophilic
Cardiovasc Interv 2008; 72:335–340. sheaths. Catheter Cardiovasc Interv 2004; 63:104–105.
74. Pancholy SB. Comparison of the effect of intra-arterial versus 94. Kozak M, Adams DR, Ioffreda MD, et al. Sterile inflammation
intravenous heparin on radial artery occlusion after transradial associated with transradial catheterization and hydrophilic
catheterization. Am J Cardiol 2009; 104:1083–1085. sheaths. Catheter Cardiovasc Interv 2003; 59:207–213.
[13/7/010/12:55:54] [67–84]
95. Tizon-Marcos H, Barbeau GR. Incidence of compartment syn- 113. Cohen M, Rentrop KP, Cohen BM. Safety and efficacy of percu-
drome of the arm in a large series of transradial approach for taneous entry of the brachial artery versus cutdown and arterio-
coronary procedures. J Interv Cardiol 2008; 21:380–384. tomy for left-sided cardiac catheterization. Am J Cardiol 1986;
96. Edmundson A, Mann T. Nonocclusive radial artery injury result- 57:682–684.
ing from transradial coronary interventions: radial artery IVUS. 114. Fergusson DJ, Kamada RO. Percutaneous entry of the brachial
J Invasive Cardiol 2005; 17:528–531. artery for left heart catheterization using a sheath. Cathet Car-
97. Burstein JM, Gidrewicz D, Hutchison SJ, et al. Impact of radial diovasc Diagn 1981; 7:111–114.
artery cannulation for coronary angiography and angioplasty on 115. Babu SC, Piccorelli GO, Shah PM, et al. Incidence and results of
radial artery function. Am J Cardiol 2007; 99:457–459. arterial complications among 16,350 patients undergoing cardiac
98. Sanmartin M, Goicolea J, Ocaranza R, et al. Vasoreactivity of the catheterization. J Vasc Surg 1989; 10:113–116.
radial artery after transradial catheterization. J Invasive Cardiol 116. Lambert M, Ball C, Hancock B. Management of acute brachial artery
2004; 16:635–638. occlusion due to trauma or emboli. Br J Surg 1983; 70:639–640.
99. Jolly SS, Amlani S, Hamon M, et al. Radial versus femoral 117. Khoury M, Batra S, Berg R, et al. Influence of arterial access sites
access for coronary angiography or intervention and the impact and interventional procedures on vascular complications after
on major bleeding and ischemic events: a systematic review cardiac catheterizations. Am J Surg 1992; 164:205–209.
and meta-analysis of randomized trials. Am Heart J 2009; 157: 118. Hildick-Smith DJ, Khan ZI, Shapiro LM, et al. Occasional-oper-
132–140. ator percutaneous brachial coronary angiography: first, do no
100. Chodor P, Krupa H, Kurek T, et al. RADIal versus femoral harm. Catheter Cardiovasc Interv 2002; 57:161–165.
approach for percutaneous coronary interventions in patients 119. Henry GA, Williams B, Pollak J, et al. Placement of an intra-
with Acute Myocardial Infarction (RADIAMI): a prospective, coronary stent via translumbar puncture. Catheter Cardiovasc
randomized, single-center clinical trial. Cardiol J 2009; 16: Interv 1999; 46:340–342.
332–340. 120. Morgan JM, Gray HH, Gelder C, et al. Left heart catheterization
101. Hetherington SL, Adam Z, Morley R, et al. Primary percutaneous by direct ventricular puncture: withstanding the test of time.
coronary intervention for acute ST-segment elevation myocardial Cathet Cardiovasc Diagn 1989; 16:87–90.
infarction: changing patterns of vascular access, radial versus 121. Walters DL, Sanchez PL, Rodriguez-Alemparte M, et al. Trans-
femoral artery. Heart 2009; 95:1612–1618. thoracic left ventricular puncture for the assessment of patients
102. Cruden NL, Teh CH, Starkey IR, et al. Reduced vascular com- with aortic and mitral valve prostheses: the Massachusetts Gen-
plications and length of stay with transradial rescue angioplasty eral Hospital experience, 1989–2000. Catheter Cardiovasc Interv
for acute myocardial infarction. Catheter Cardiovasc Interv 2007; 2003; 58:539–544.
70:670–675. 122. Sorajja P, Cabalka AK, Hagler DJ, et al. Successful percutaneous
103. Louvard Y, Benamer H, Garot P, et al. Comparison of transradial repair of perivalvular prosthetic regurgitation. Catheter Cardio-
and transfemoral approaches for coronary angiography and vasc Interv 2007; 70:815–823.
angioplasty in octogenarians (the OCTOPLUS study). Am J 123. Ommen SR, Higano ST, Nishimura RA, et al. Summary of the
Cardiol 2004; 94:1177–1180. Mayo Clinic experience with direct left ventricular puncture.
104. Hildick-Smith DJ, Walsh JT, Lowe MD, et al. Transradial coro- Cathet Cardiovasc Diagn 1998; 44:175–178.
nary angiography in patients with contraindications to the fem- 124. Hind D, Calvert N, McWilliams R, et al. Ultrasonic locating devices
oral approach: an analysis of 500 cases. Catheter Cardiovasc for central venous cannulation: meta-analysis. BMJ 2003; 327:361.
Interv 2004; 61:60–66. 125. Augoustides JG, Horak J, Ochroch AE, et al. A randomized con-
105. Louvard Y, Kiemeneij F. Available at http://www.tctmd.com/ trolled clinical trial of real-time needle-guided ultrasound for
txshow.aspx?tid=1326&id=60870&trid=2. Accessed 10/27/2009. internal jugular venous cannulation in a large university anesthe-
106. Cooper CJ, El Shiekh RA, Cohen DJ, et al. Effect of transradial sia department. J Cardiothorac Vasc Anesth 2005; 19:310–315.
access on quality of life and cost of cardiac catheterization: a 126. Ruesch S, Walder B, Tramer MR. Complications of central
randomized comparison. Am Heart J 1999; 138:430–436. venous catheters: internal jugular versus subclavian access–a
107. Kiemeneij F. Cost-effectiveness of transradial coronary access. systematic review. Crit Care Med 2002; 30:454–460.
J Invasive Cardiol 2007; 19:354. 127. Wang DS, Chu LF, Olson SE, et al. Comparative evaluation of
108. Mann T, Cowper PA, Peterson ED, et al. Transradial coronary noninvasive compression adjuncts for hemostasis in percutane-
stenting: comparison with femoral access closed with an arterial ous arterial, venous, and arteriovenous dialysis access proce-
suture device. Catheter Cardiovasc Interv 2000; 4:150–156. dures. J Vasc Interv Radiol 2008; 19:72–79.
109. Roussanov O, Wilson SJ, Henley K, et al. Cost-effectiveness of the 128. Zahn R, Fromm E, Thoma S, et al. Local venous thrombosis after
radial versus femoral artery approach to diagnostic cardiac cardiac catheterization. Angiology 1997; 48:1–7.
catheterization. J Invasive Cardiol 2007; 19:349–353. 129. Coto HA. Closure of the femoral vein puncture site after trans-
110. Mann JT III, Cubeddu G, Arrowood M. Operator radiation catheter procedures using Angio-Seal. Catheter Cardiovasc
exposure in PTCA: comparison of radial and femoral approaches. Interv 2002; 55:16–19.
J Invasive Cardiol 1996; 8(suppl D):22D–25D. 130. Mylonas I, Sakata Y, Salinger M, et al. The use of percutaneous
111. Proudfit WL, Shirey EK, Sones FM Jr. Selective cine coronary suture-mediated closure for the management of 14 French fem-
arteriography. Correlation with clinical findings in 1,000 patients. oral venous access. J Invasive Cardiol 2006; 18:299–302.
Circulation 1966; 33:901–910. 131. Bagai J, Zhao D. Subcutaneous “Figure-of-Eight” stitch to achieve
112. Campeau L. Percutaneous brachial catheterization. Cathet Car- hemostasis after removal of large-caliber femoral venous sheaths.
diovasc Diagn 1985; 11:443–444. Cardiac Interv Today 2009; 5:22–23.
[2/7/010/20:36:32] [85–97]
Right heart catheterization

Franz R. Eberli
INTRODUCTION The PA catheter (Fig. 9.1) is advanced through the right

In addition to Doppler echocardiography and cardiac magnetic atrium, the right ventricle (RV), and the PA until a PA occlusion
resonance imaging right heart catheterization is a complemen- pressure is reached (Fig. 9.2). The static column of blood
tary, but indispensable investigation for hemodynamic assess- between the tip of the PA catheter and the pulmonary vein
ment and diagnosis of many cardiac diseases. Right heart will transmit the pressure from the left atrium (LA). During
catheterization provides data on the pressures and oxygen diastole, when the mitral valve is open, the measured pressure
saturations in the right heart chambers and the pulmonary also corresponds to the left ventricular diastolic pressure
artery (PA), including PA occlusion pressure or “wedge” pres- (Fig. 9.2). A prerequisite of a correct measurement is that the
sure. Combined with the data obtained during simultaneous pulmonary venous pressure exceeds pulmonary alveolar pres-
left heart catheterization cardiac output, pulmonary and sys- sure. This is more likely the case, when the catheter tip is directed
temic vascular resistances and ejection fractions are calculated, into the lower lobe (8). The lung tissue between the tip of the
and shunt detection and quantification in structural heart dis- catheter and the left heart results in a damping (2–4 mmHg) and
ease can be performed. From these measurements information delaying (100–150 milliseconds) of the pressure wave in the
about preload, afterload, and contractility are derived. Hemo- pulmonary capillary wedge pressure (PCWP) as compared
dynamic responses to changes in loading conditions and/or to with left ventricular or left atrial pressure (LAP) (Fig. 9.3).
pharmacologic interventions are often performed for accurate Cardiac output measurements are usually performed by
evaluation of the physiology of a specific condition and are part thermodilution, preferable in a normal and high output states,
and parcel of right heart catheterizations in the cardiac catheter- or by the Fick method, which is preferable in low output states,
ization laboratory. This information is particularly valuable in valvular regurgitation or intracardiac shunts. Shunt detection
the assessment of adult congenital heart disease and pulmo- and quantification is nowadays complementary to echocardiog-
nary hypertension prior to any therapeutic intervention or raphy results. Echocardiographic findings should be reviewed
operation (1,2). before performing right heart catheterization, so that the inva-
In contrast to its undisputed value as diagnostic investiga- sive procedure can be tailored toward the unresolved and
tion in the cardiac catheterization laboratory, the value of bedside specific questions. This will allow the investigator to shorten
use of a PA catheter in critically ill patients is more controversial. the oxymetry run and help to plan a potentially valuable
Reports that its use leads to increased mortality and morbidity (3) pharmacologic intervention or volume load.
have resulted in a careful evaluation of the use of bedside right
heart catheters in patients with cardiac disease (4). Recently, two
randomized trials (PAC-MAN and ESCAPE) have found no INDICATION
increased mortality with the use of PA catheters in critically ill Right heart catheterization is no longer part of every diagnostic
patients and have concluded that PA catheters are safe for use in heart catheterization. Echocardiography, other imaging modal-
an appropriate patient populations (5,6). ities and noninvasive hemodynamic measurements have
reduced the need for right heart catheterization. Nevertheless,
direct measurements of pressure, flow and oxygen saturations
ANATOMIC CONSIDERATIONS are often necessary to correctly diagnose or quantify cardiac
AND FUNDAMENTALS diseases. For example, despite the undisputable achievements,
Access to the right heart can be gained via the inferior or Doppler echocardiography has its limitations for correct assess-
superior vena cava. In the cardiac catheterization laboratory ment of pressures (9). The indications for right heart catheter-
the inferior vena cava is the preferred approach, for bedside ization can be divided into two main categories: (A) diagnostic
right heart catheterization the superior approach is used. The catheterization for establishing a diagnosis and for planning
superior vena cava is assessed either from the (right) internal and guiding a therapeutic intervention and (B) monitoring and
jugular vein or the (left) subclavian vein. Although in the guiding of intensive medical care or perioperative hemody-
catheterization laboratory most right heart catheters are per- namics (Table 9.1) (10).
formed via femoral access, certain conditions make a superior
caval vein approach preferable. Such conditions are: Suspected
femoral vein/iliac vein thrombosis, renal vein thrombus, infe- CONTRAINDICATION
rior vena cava filter, anomalous inferior vena cava. Other Absolute contraindications to right heart catheterizations are
conditions, such as massive dilation of the right sided cham- mechanical tricuspid or pulmonic valve prosthesis and terminal
bers, severe tricuspid or pulmonary regurgitation, and pulmo- illness. Relative contraindications are endocarditis, tumor or
nary hypertension are technically easier to assess by a superior thrombus in the right heart chambers, and newly implanted
vena cava approach. pacemaker or defibrillator electrodes. A profound coagulopathy
[2/7/010/20:36:32] [85–97]
Figure 9.1 Diagram depicting the PA catheter. A PA catheter with two pressure monitor lumens is depicted. The distal lumen goes to the tip
of the catheter to measure PA pressure and, if the balloon is inflated and the PA occluded, the pulmonary capillary wedge pressure is
measured. A thermistor near the catheter tip measures PA blood temperature and is used for thermodilution cardiac output measurements.
The proximal lumen is located 30 cm from the tip of the catheter and usually lies within the right atrium. This lumen is also used for injection of
cold saline for thermodilution cardiac output measurements. In the cross section the distribution of the four lumens within the catheter are
depicted. Source: From Ref. 7, with permission. Abbreviation: PA, pulmonary artery.
Figure 9.3 Simultaneous left ventricular and PCWP. This figure

depicts the simultaneous pressure measurement with a tip mano-
Figure 9.2 Principles of right heart catheterization. The balloon- meter catheter (Millar catheter) in the left ventricle and in pulmonary
tipped right heart catheter is inserted from an inferior vena cava capillary wedge position. The left atrial contraction results in an a-
approach into the right atrium (RA), through the tricuspid valve into wave that is transmitted into the LV and retrogradely into the PCWP
the right ventricle (RV) and up into the PA and “wedged” into the [the time delay (Dt) is usually 80–140 milliseconds]. Similarly the v-
distal PA. Distal to the inflated balloon a blood column is present wave is delayed and dampened (2–4 mmHg). After mitral valve
between the PV and the catheter. The PV is connected to the LA opening (MVO) the v-wave should decrease simultaneously with the
and, importantly, if the MV is open, to the LV. According to the law of LV pressure. The recorded delayed pressure decay is the result of
pressures in communicating vessels the left ventricular end diastolic the dampening of the pressure transmission through the lung tissue.
pressure corresponds to left atrial pressure, which in turn corre- Abbreviations: PCGLV, phonocardiogram left ventricle; PCGPA, pho-
sponds to PV pressure and pulmonary capillary wedge pressure. nocardiogram pulmonary artery; LVP, left ventricular pressure;
Abbreviations: PA, pulmonary artery; PV, pulmonary vein; LA, left PCWP, pulmonary capillary wedge pressure; ECG, electrocardio-
atrium; LV, left ventricle; MV, mitral valve. gram; LV, left ventricle.
[2/7/010/20:36:32] [85–97]
RIGHT HEART CATHETERIZATION 87
Table 9.1 Indications for Right Heart Catheterization

A. Diagnosis and planning of therapeutic interventions
1. Valvular heart disease
Assessment of severity of valve disease and concomitant pulmonary hypertension.
Planning of surgigal or interventional valve interventions.
2. Intracardiac shunts
Shunt detection and quantification. Assessment of concomitant pulmonary hypertension. Exploratory balloon occlusion of the defect.
Vasoreactivity test in case of pulmonary hypertension.
3. Left heart failure
To differentiate between cardiogenic or noncardiogenic pulmonary edema. Guide therapy in acute and chronic heart failure.
Differentiate between diastolic and systolic dysfunction. Test for reversible pulmonary hypertension in transplant candidates.
4. Shock states
Differentiate shock states. Cardiogenic vs. noncardiogenic, hypovolemic, septic shock.
5. Acute myocardial infarction
Complicated by hypotension, unclear volume status (acute MI and bleeding, acute MI and renal insufficiency). Right ventricular
infarction. Mechanical complications (ventricular septal defect, papillary muscle rupture).
6. Pulmonary hypertension
Diagnostic gold standard. Assessment of etiology of pulmonary hypertension.
Determination of severity. Testing for vasoreactivity.
7. Pericardial diseases
Differentiation between constrictive and restrictive physiology. Cardiac tamponade (only when echocardiography is unavailable or
nondiagnostic).
B. Monitoring of intensive medical therapy
1. Heart failure
Guidance of vasodilator, positive inotrope and diuretic therapy, perioperative management of patients with decompensated heart
failure.
2. Myocardial infarction
Guidance of pharmacologic and mechanical support. Management of pulmonary edema that does not respond to usual medical
treatment.
3. Perioperative use in cardiac surgery
To determine etiology of low cardiac output, differentiate between left and right heart failure, management of pulmonary hypertension.
4. Severe acute respiratory distress syndrome
Assess cardiac output during positive end-expiratory pressure trials.
with an INR >2 or low platelet count (<20,000/mm3 ) increases thermal input to produce a thermodilution washout curve from
the bleeding risk, particularly when a jugular vein or subclavian which cardiac output is calculated.
vein approach is chosen. Periprocedural administration of fresh The soft Swan-Ganz catheter has a low risk of injury.
frozen plasma or platelets might be considered under certain However, when floated from a femoral vein it is not easily
circumstances. guided by torquing. Maneuverability is improved by guide-
wires (usually 0.021–0.032 in). Unfortunately, the soft catheter
with a small diameter has a poor frequency response which
TECHNIQUE might impede pressure measurements (11). Instead of Swan-
Equipment Ganz catheters multipurpose catheters or stiffer (woven Dacron)
The most widely used and safest catheter for right heart Goodale-Lubin or Cournand catheters might be used in the
catheterization is a multilumen, balloon-tipped flotation cathe- cardiac catheterization laboratory. If a PA or pulmonary vein
ter, the so-called Swan-Ganz catheter (Fig. 9.1). When used angiography is to be performed, a regular pigtail catheter or an
from a superior caval vein approach the catheter floats easily angled (Grolman) pigtail catheter is the preferred catheter. A
through the right heart chambers and the balloon wedges into Berman catheter can also be used. This is a balloon-tipped
the distal PA. The distal and proximal lumen of the catheter catheter specifically designed for right-sided angiography. It
allow measurement of pressures in the PA and in the right has no end-holes but several side-holes proximal to the balloon.
atrium, the built-in thermistors allow measurements of cardiac The inflated balloon stabilizes the catheter and the large lumen
output by thermodilution. The standard PA catheter is 110 cm allows flows similar to that in pigtail catheters. The flow and
long and has a distal and a proximal lumen (ending 30 cm from injection rate on the right side should be somewhat lower than
the catheter tip) that in most patients will lie within the right on the left side and the maximal pressure of injection should be
atrium, plus a thermistor filament near the tip. For diagnostic reduced to 600 psi.
purposes in the catheterization laboratory double-lumen bal-
loon-tipped catheters without proximal lumen and without
thermistors are often employed. In the CCU and ICU more Access Site and Venous Introducer Insertion
sophisticated catheters with a built-in oxygen probe and con- The Femoral Vein
tinuous cardiac output measurements are used. In these eight The femoral vein is the preferred access when the right heart
French catheters, a thermal filament in the RV heats the catheterization is part of a diagnostic or interventional proce-
surrounding blood (7.5 W energy transformed to heat the dure in the catheterization laboratory. For a description of
blood to 39–438C). The heat change at the thermistor near the gaining access and inserting the sheath see chapter 8. The
tip of the catheter is cross correlated with the right ventricular advantages of the femoral access are the ease of cannulation
[2/7/010/20:36:32] [85–97]
and the ease of compression, as well as the absence of the risk advance the tip of the catheter into a wedge position in the
of a pneumothorax. Disadvantages are the increased technical distal PA.
difficulties of guiding and advancing the Swan-Ganz catheters In case of an enlarged right atrium direct advancement of
as well as the risk of a femoral artery puncture. For longer term the PA catheter via the tricuspid valve might not be easily
monitoring there are additional disadvantages to the femoral achieved. It is sometimes necessary to increase the bend on the
vein access, particularly patient immobility, the higher risk of catheter. To do that one can loop the catheter by bending it
thrombosis and infection. against the lateral wall of the right atrium or by engaging it into
the ostium of the hepatic vein. The catheter is then advanced
Internal Jugular Vein and rotated clockwise. This will help to advance the catheter
If the internal jugular vein access is used, the right side is the into the right atrium. Another helpful technique is to loop the
preferred side, since it provides a direct access to the right catheter toward the lateral wall of the right atrium with
atrium. Furthermore, the left internal jugular vein sometimes the balloon deflated. Subsequently, the operator can advance
has a venous valve at the entrance into the subclavian vein. The the catheter such that a loop in the right atrium is formed.
advantages of a jugular vein approach are the easy compressi- When the tip of the loop is directed toward the tricuspid valve
bility of the access site and the low risk of a pneumothorax. The the balloon is reinflated and the catheter floats across the
disadvantage is the risk of puncture of the carotid artery. The tricuspid valve, similar to floating the balloon catheter from
internal jugular vein is the preferred access for hemodynamic the superior vena cava approach. If the tip of the catheter
monitoring in the ICU or CCU and perioperatively. In the reaches the right ventricular outflow tract, it can usually be
cardiac catheterization laboratory it is often preferred when advanced easily into the PA and the wedge position. Occasion-
simultaneously a myocardial biopsy (e.g., in posttransplant ally, the balloon is so soft that it will not advance despite a nice
patients) is performed or when inferior access is impossible. loop in the RA across the tricuspid valve and into the RV.
Introducing a guidewire into the loop, but not to the tip of the
Subclavian Access catheter might help to advance it further into the PA.
The great advantage of subclavian access is the patient comfort Catheters advanced from the vena cava inferior tend to
and the ease of insertion of the catheter. The disadvantages are advance into the left PA, whereas catheters advance from the
the danger of puncture of the subclavian artery and the risk of a vena cava superior tend to advance into the right PA. If both
pneumothorax. Subclavian access is seldom used in the cathe- pulmonary arteries need to be engaged for a procedure, a
terization laboratory, it is mainly reserved for long-term hemo- guidewire can be introduced into the PA catheter with an
dynamic monitoring in the CCU and ICU. appropriate bend so that the catheter can be advanced into
both pulmonary arteries. It is often necessary to use the stiff
Pulmonary Artery Catheter Insertion end of the guidewire to assure an appropriate bend. In such a
After flushing, the Swan-Ganz catheter is introduced into the case, the guidewire is never to be advanced to the tip of the PA
sheath and brought up into the common iliac vein, where the catheter. The stiff end of a guidewire has a very high risk of a
balloon is inflated under fluoroscopy. An inflation of the bal- perforation and should never be exposed into the right heart or
loon in a small vein should be avoided. When the balloon is not in the PA. Even the soft end of the guidewire carries a certain
inflated without fluoroscopic control, it has to be attached to risk of perforation in the PA and should not be routinely used.
the pressure manifold, so that the balloon may be deflated After recording the pulmonary wedge pressure, the bal-
immediately if a pressure increase, that is an over-inflation is loon is deflated under slight tension, such that no abrupt
detected. During advancement deviation from the straight path forward movement of the catheter tip occurs. Then the PA
along the spine usually suggests entry into a renal or hepatic pressure is measured and blood oxygen saturation samples are
vein. The catheter is then withdrawn, rotated and advanced taken simultaneously from the PA and from a systemic artery
further. When the right heart catheter is performed to detect or (femoral or radial) to measure cardiac output according to the
quantify a left-to-right shunt, the catheter should first be Fick formula. Pressure measurements (and if needed oxygen
advanced into the vena cava superior. To do this the balloon saturations) are recorded during withdrawal of the catheter
is best deflated and the catheter rotated to the lateral wall and from the PA to the RV and right atrium by counterclockwise
than advanced by further counterclockwise rotation movement. rotation and pulling.
It is advisable to transverse the straight path from the vena cava Advancing a multipurpose catheter from the femoral
inferior to superior by using a guidewire within the Swan-Ganz vein to the PA is done by the same technique. In the right
catheter. Once the catheter is advanced into the vena cava, a atrium the guidewire is withdrawn and the tip of the multi-
blood sample is drawn from the vena cava superior and at the purpose catheter is directed medially toward the tricuspid
entrance of the vena cava superior into the right atrium (for valve and advanced into the RV. Under clockwise rotation it
detection or exclusion of anomalous pulmonary venous return). is advanced into the RV and up to the PA. The guidewire is
The advancement of the PA catheter from the right then advanced into the PA and the catheter gently brought
atrium into the RV and PA is usually done by passing directly, forward into a wedge position.
that is medially across the tricuspid valve into the RV. This When advancing the catheters into the right heart ana-
directs the tip of the catheter toward the apex of the RV. tomical variants and anomalies, such as patent foramen ovale, a
Therefore, the catheter has to be pulled back slightly and persistent left superior vena cava, and anomalous returning
than rotated further clockwise. To transmit the torque, it is pulmonary veins, might direct the catheter into an unwanted
important to maintain pressure on the catheter and use slight cavity. Fluoroscopy and oxymetry help to correctly locate
forward and backward movements. When the balloon tip is the position of the catheter. Erroneous catheter placement into
pointing upward toward the right ventricular outflow tract, the the LA and pulmonary vein is recognized by a path through the
catheter is advanced quickly into the PA. Deep inspiration heart and deviation into a posterior position on a lateral view
might help to advance the catheter up into the PA and to and can be confirmed by oxygen saturation measurements. A
[2/7/010/20:36:32] [85–97]
catheter position in the coronary sinus is suspected by fluoros- The most serious complication of the PA catheter is PA
copy and confirmed by oxygen saturation measurements rupture. In a prospective study the rate of PA rupture was
between 20% and 30%. 0.2%, in retrospective studies it varied between 0.001% to
Right heart catheterization from a superior caval vein 0.47% (14,18). Elderly patients and patients on chronic steroid
approach is usually done with an inflated balloon and the therapy are at increased risk, whereas pulmonary hyperten-
balloon connected to the pressure transducer. The distance sion does not constitute a higher risk for this complication
marker on the catheter (every 10 cm) is helpful for orientation. (19,20). The rupture may cause an asymptomatic pseudoa-
After 15 to 20 cm, the right atrial tracing is seen and with the neurysm, hemoptysis, that stops spontaneously, or hemopty-
inflated balloon the catheter is advanced across the tricuspid sis and hemothorax (18). In case of hemoptysis and
valve. The catheter is then advanced into the PA and into hemothorax the mortality rate is reportedly between 45%
wedge position. This position should be reached after about and 70%. Diagnosis should be suspected upon the onset of
50 to 55 cm. If no wedge position is reached after this distance, chest pain related to catheter advancement or balloon infla-
the catheter might be coiled in the RV. It is advisable to deflate tion, frank hemoptysis and dyspnea. If frank hemoptysis is
the balloon, withdraw the catheter into the right atrium, and present, immediate selective intubation of the uninjured lung
start the process all over. should be performed. If no hemothorax is present, techniques
such as balloon tamponade of the ruptured artery, tamponade
of the bleeding lung by bronchoscopy, and embolization of the
Complications ruptured artery have all been used to control the bleeding (21).
The complications of PA catheters can be divided into compli-
In case of hemothorax the treatment is mostly surgical and a
cations related to vascular access, to catheter insertion and to
lobectomy or pneumectomy is sometimes the only option to
catheter residence (Table 9.2) (12). Additional complications are
stop the bleeding (22). To avoid this most serious complication
related to vasoreactivity testing, volume loading, and pulmo-
the balloon should only be inflated in the large proximal PA,
nary or right ventricular angiography (see chaps. 10 and 22)
the time of the balloon in wedge position should be minimized
(13). The risk of catheter related complications seems to have
and traction on the catheter when deflating the balloon should
declined over time. In earlier reports an approximately 5% rate
be maintained to prevent a rapid forward movement of the
of serious complications was reported (14–16). In recent studies
catheter tip (18).
serious complications were encountered in less than 1%, even
Insertion of PA catheters in patients with preexisting left
in patients with pulmonary hypertension (5,13). Hemodynamic
bundle branch block is complicated in approximately 3% to 5%
monitoring in critically ill patients carries additional risks
by the occurrence of additional right bundle branch block that
related to medium to long-term indwelling of the PA catheter.
can lead to complete heart block (15,16). Therefore, when
Thromboembolic complications and infections are the most
performing a PA catheter in a patient with a left bundle branch
frequent complications encountered in these patients (17). In
block a temporary pacemaker should be available for emer-
the ESCAPE trial 4.2% of patients suffered PA catheter related
gency pacing if needed.
complications, of which two thirds were attributed to long-term
catheter residence. These complications were infections and
pulmonary infarction/haemorrhage (6).
PRESSURE MEASUREMENTS
AND PRESSURE WAVE FORMS
Table 9.2 Complications of PA Catheter A prerequisite for accurate pressure measurements is a pres-
sure recording with a satisfactory frequency response and few
Vascular access complications artefacts. The most common problem in pressure recording on
– Vasovagal reaction the right side is related to the quality of the soft catheters. The
– Arterial puncture long soft tubing results in over-damping of the pressure wave
– Arteriovenous fistula
forms similar to air bubbles or to kinking in the catheter. Stiffer
– Bleeding from insertion site
– Nerv injury catheters and hyperdynamic states result in under-damping.
– Air embolism Whipping the catheter against the septum results in pressure
– Pneumothorax, hemothorax (subclavian, internal jugular artefacts. Therefore, it is important to carefully examine the
vein approach) position of the pressure transducer, the quality of the wave
Related to catheter insertion form and the simultaneously recorded electrocardiogram. A
– Arrhythmias (supraventricular tachycardia, ventricular crisp dicrotic notch on the PA pressure usually indicates a
premature beats, VT, VF) properly responsive pressure system.
– Right bundle branch block or complete heart block The intracardiac and transmural pressures are greatly
– Injury to chordae in right ventricle influenced by intrathoracic pressures. Intrathoracic pressure is
– Tricuspid regurgitation
transmitted to the intracardiac pressures and thus pressures
– Dislodgement of pacemaker leads
– PA rupture/right ventricular perforation will vary with respiration. In normal physiology inspiration
Related to catheter residence will decrease intrathoracic pressure and increase venous
– PA rupture return. The right atrium and ventricle being rather elastic will
– Pulmonary infarction accommodate this increased volume without greatly increasing
– Thrombosis intracardiac pressure. Thus in normal hearts the net effect of
– Infection/endocarditis/thrombophlebitis inspiration is a decrease in right-sided pressures and an
– Balloon rupture / embolization increase during expiration (23). At end-expiration the intra-
Abbreviations: PA, pulmonary artery; VT, ventricular tachycardia; thoracic pressure is almost zero and closest to atmospheric
VF, ventricular fibrillation. pressure. Therefore, all pressures should be recorded at
[2/7/010/20:36:32] [85–97]
end-expiration. For a nice steady recording ask the patient to

stop breathing at the end of a nonlabored expiration. Intra-
thoracic pressures are reversed in a mechanically ventilated
patient. On a mechanical ventilator intrathoracic pressures are
increased and venous return impaired during inspiration, and
conversely intrathoracic pressures are decreased and venous
return increased during expiration. However, as is the case in
normal respiration, the intrathoracic pressures are closest to
atmospheric pressure at the end of expiration and should be
measured at this point, too (23).
The normal pressure wave forms are depicted in Fig-
ures 9.3 to 9.6. If there is an impairment to filling (e.g., restric-
Figure 9.4 Right atrial pressure. The right atrial systole follows the tive physiology) or a great increase in right heart filling (e.g., in
p-wave on the electrocardiogram. The a-wave indicates atrial con- decompensated heart failure) the pressures will not vary with
traction and is followed by the x descent that corresponds to atrial respiration. Right atrial pressures may remain flat (and ele-
relaxation. The closure of the tricuspid valve produces a slight vated) throughout the respiratory cycle or even increase during
upward deflection of the x descent and is called the c-wave. The inspiration (Kussmaul’s sign) (Fig. 9.7).
x0 descent represents the descent of the atrioventricular ring during The PCWP is particularly sensitive to pathologic intra-
ventricular systole and atrial relaxation. The c-wave is not always thoracic pressures. Since PCWP is an indirect measurement of
present and the wave form is than reduced to the a-wave and the x LAP, transmitted by a blood column from the LA to the tip of
descent. Following the t-wave on the ECG a v-wave is present. The the catheter, any changes in intrathoracic pressures that influ-
v-wave represents the atrial filling during diastole. The v-wave is
ence this fluid column will change PCWP. An increase in
followed by the y descent that marks the opening of the tricuspid
valve and the emptying of the atrium during the rapid diastolic intraalveolar pressure (e.g., in chronic obstructive lung disease,
ventricular filling. Abbreviation: RA, right atrium. Figures 9.4 to 9.6 positive end-expiratory pressure) will lead to an overestimation
kindly provided by H. Hirzel, MD. of left atrial filling pressures.
Figure 9.5 RVP and PAP. The RVP and PAP are simultaneously recorded by a tip manometer (Millar catheter). Also recorded are the
phonocardiogram (PCG) and the ECG. The patient has a minimal pulmonic stenosis and a mild pulmonary regurgitation. The atrial contraction
results in an a-wave that occurs after the p-wave of the ECG. The end of the a-wave marks closure of the tricuspid valve as evidence by the
first heart sound and represents right ventricular end diastolic pressure. This is followed by ventricular systole. Ventricular relaxation results in
closure of the pulmonary valve ¼ second heart sound (2. HS) and isovolumic relaxation of the RV until tricuspid valve opens and rapid
diastolic filling of the RV occurs. Usually right ventricular end diastolic pressure and peak systolic pressure are measured. The PAP contains a
v-wave, which corresponds to the right ventricular systolic pressure and relaxation. The pulmonic valve closure produces a dictrotic notch.
The decline of the pressure tracing continues throughout diastole until PAP starts to increase again with ventricular contraction.
Abbreviations: ECG, electrocardiogram; RV, right ventricle; RVP, right ventricular pressure; PAP, pulmonary artery pressure.
[2/7/010/20:36:32] [85–97]
Figure 9.6 Simultaneous LV and LAP. Depicted are

the pressure wave forms recorded by fluid-filled
catheters in the LV and the LA. The LAP is initially
recorded as mean pressure and then in a phasic
condition. The a-wave of the LA contraction is fol-
lowed by the x descent that marks the descent of the
atrioventricular ring during ventricular systole and
atrial relaxation. The filling of the LA produces the
v-wave and the rapid emptying of the LA after mitral
valve opening (MVO) results in the y descent. Atrial
and left ventricular pressures are overlapping during
diastasis of diastole. The a-wave of atrial contraction
is transmitted into the LV and results in the a-wave of
the LV pressure tracing. The trough after the a-wave
marks left ventricular end diastolic pressure. Unchar-
acteristically, in this example the a-wave is larger
than the v-wave. Normally in LAP the v-wave is larger
than the a-wave. Abbreviations: LA, left atrium;
LV, left ventricle; ECG, electrocardiogram; LVP, left
ventricular pressure; LAP, left atrial pressure.
Figure 9.7 Kussmaul sign. The right atrial pressure

tracing of a patient with restrictive cardiomyopathy is
depicted. Right atrial pressure is elevated. The prom-
inent x and y descents give the wave form the char-
acteristic M oder W shaped appearance. During
inspiration right atrial pressure increases and the y
descent is further augmented. This paradoxical
increase in right atrial pressure during inspiration is
called the Kussmaul sign.
CARDIAC OUTPUT MEASUREMENTS therefore become common practice to normalize cardiac output
The delivery of the blood to the body per minute is termed to BSA as the so-called cardiac index (L/min/m2 BSA). In the
cardiac output. Cardiac output is dependent on metabolic state, cardiac catheterization laboratory cardiac output is most often
body size, age, and a number of other factors such as posture, determined by the Fick oxygen consumption method or the
anxiety, and body temperature (24). Baseline cardiac output is thermodilution technique. In addition, angiographic cardiac
mainly determined by the metabolic rate. The baseline meta- output measurements (stroke volume heart rate) might be
bolic rate is closely correlated to body surface area (BSA). It has used.
[2/7/010/20:36:32] [85–97]
Fick Principle curve is measured and this area is used to calculate cardiac
The Fick principle states that the total uptake or release of a output. The area under the curve is inversely related to cardiac
substance by any organ is the product of the arteriovenous output. Because the body acts as large heat sink, there is no
concentration difference of the substance and the blood flow to recirculation. Hence successive measurements can be made
that organ (24). In the lungs the released substance is oxygen without waiting between measurements. Usually, five meas-
and the pulmonary blood flow can be calculated from the urements are performed. The lowest and the highest measure-
measured differences in arterial and venous oxygen saturations ment discarded and the mean of the three remaining
and the oxygen uptake by the lungs per minute. In the absence measurements recorded as cardiac output.
of intracardiac shunts pulmonary flow equals systemic flow. Apart from proper calibration of the different catheters,
The arteriovenous oxygen gradient is calculated from the correct onset of recording and checking the correct position of
difference in oxygen saturation between arterial and mixed the catheters, two additional main limitations of the thermodil-
venous blood. Mixed venous blood is best obtained from the uation technique have to be considered. Commercially avail-
PA or alternatively from an RV blood sample. For correct able systems are designed to prevent warming of the injectate
measurement of blood oxygenation in the lungs a pulmonary by the hands of the operator. However, warming of the cold
venous sample should be measured. For practical reasons a solution caused by prolonged stay of the injectate in the right
sample from the LV or the femoral artery is taken. The small atrium secondary to severe tricuspid regurgitation or low car-
decrease in saturation due to bronchial and thebesian vein diac output makes the thermodilution method unreliable (27).
drainage is negligible. In the arterial and venous samples Similarly, heavy respiration might induce PA temperature
oxygen saturation is measured. The calculation of the oxygen changes and may result in inaccuracies of thermodilution car-
content assumes that red cells with 100% oxygen saturation diac output measurements (23).
carry 1.36 mL O2/g hemoglobin. Oxygen content (mL O2/L
blood) ¼ % oxygen saturation 1.36 (mL of O2/g hemoglobin)
hemoglobin (g/dL) 10 (converts from dL to L). VASCULAR RESISTANCE
Oxygen consumption can be measured by a metabolic Ohm’s law defines hydraulic resistance as the ratio of mean
rate meter (polarographic cell method) where the patient pressure drop (DP) across the hydraulic system to laminar flow
breathes ambient room air in a steady state for several minutes through it. In such a system the resistance to flow depends
and the metabolic rate meter gives a readout of oxygen con- solely on the diameter of the tubing and the viscosity of the
sumption in liters per minute. Alternatively, basal oxygen fluid. The concept of vascular resistance was established in
consumption can be assumed from BSA, age, and sex. In analogy to Ohm’s law. However, the cardiovascular system
young patients a basal oxygen consumption of 3 mL/kg body differs in a number of ways from a hydraulic system: for
weight or 125 mL/m2 BSA is used in many laboratories. For example, it is a pulsatile system; the blood is an inhomoge-
elderly patients an oxygen consumption of 110 mL/m2 is neous fluid, the vessels are elastic; and pressure waves are
normal. It is important, however, to understand the limitations reflected. Therefore, the concept of vascular resistance is ques-
of this assumption. Baseline oxygen consumption can vary by tionable. Vascular impedance, defined as the ratio of pulsatile
as much as 25% between patients (25). If the patient is not in a pressure to pulsatile flow, would be a more accurate concept to
steady state because of anxiety, dyspnea, tachycardia, or is over describe the resistance and elasticity of the cardiovascular
sedated and breathes shallowly oxygen consumption is either system (24). Nevertheless, the concept of systemic and pulmo-
increased or decreased to an unpredicted level. The error is nary vascular resistance has gained wide acceptance. Its use-
greatest in tachycardic patients. Caution is therefore warranted fulness to assess pathophysiologic conditions has been
when cardiac output is calculated using an assumed oxygen established by many empiric data. Calculation of vascular
consumption, and consequently also when this value is used to resistances is now widely used for clinical decision making.
calculate valve areas or vascular resistances. Systemic or pulmonary vascular resistance might also be
Another important source of error in calculating cardiac termed systemic or pulmonary arteriolar resistance. However,
output by the Fick method is the administration of supplemen- only about 60% of the pressure drop across the system occurs at
tal oxygen to the patient. This makes it almost impossible to the arteriolar level, the other 40% at other sites in the vascular
calculate oxygen content (dissolved oxygen in plasma!) and bed (24). Therefore, systemic and pulmonary vascular resis-
hence cardiac output. Therefore, supplemental oxygen should tance is the preferred term.
be discontinued at least 15 minutes before determination of The systemic vascular resistance is calculated as the ratio
cardiac output by the Fick oxygen consumption technique. of the pressure drop from aorta to right atrium (mean aortic
pressure—mean right atrial pressure) to systemic flow (Qs). The
result of this ratio is expressed either in an arbitrary resistance
Thermodilution Technique unit (mmHg/L/min), also called hybrid resistance unit or
The thermodiluation technique to measure cardiac output is an Wood unit (according to its first introduction by Dr Paul
indicator dilution technique, with a cold fluid as the indicator. Wood). This hybrid resistance unit is used in pediatric cardi-
It is based on the Stewart-Hamilton-Principle: a known quan- ology. In adult cardiology it is usually converted into metric
tity of an indicator is injected as bolus into the circulating resistance unit, expressed in dynes sec cm5 by use of the
system and after thorough mixing a concentration-time curve is conversion factor 80 (Table 9.3). In pediatric cardiology the
sampled downstream (26). Practically, cold saline is injected via resistance is also usually normalized for BSA, thus resulting in
the proximal lumen of the PA catheter (Fig. 9.1) and the a vascular resistance index. The resistance index is obtained by
transient drop in blood temperature is sensed by the thermistor dividing the mean pressure drop by the cardiac index (not
at the distal end of the catheter. The cardiac output computer obtained by dividing the resistance by BSA).
generates a thermodilution curve by plotting the decline in Pulmonary vascular resistance is calculated as the ratio of
temperature (8C) versus time (seconds). The area under the the pressure drop from the PA to the LA to pulmonary flow (Qp).
[2/7/010/20:36:32] [85–97]
Table 9.3 Formulas for Calculation of Cardiac Output and Vascular Resistances
Oxygen content ¼ O2 saturation 1:36 Hgb 10 ¼ mL=L
Cardiac output (CO) by the Fick oxygen consumption method is
O2 consumption ðmL=minÞ
CO ¼
AVO2 difference ðmLO2 =100 mL bloodÞ 10
Wt 3 mL O2 =kg BW
CO ¼ ¼ L=min
ðAO2 % VO2 %Þ 1:36 Hgb 10
CO
Cardiac index ðCIÞ ¼ ¼ L=min=m2
BSA
COðmL= minÞ
Stroke volume ðSVÞ ¼ ¼ mL=beat
Heart rate ðbeats=minÞ
SVðmL=beatÞ
Stroke index ðSIÞ ¼ ¼ mL=beat=m2
BSAðm2 Þ
Transpulmonary pressure gradient ¼ mean PAP mean PCWP ¼ mmHg
mean PAP mean PCWP
Pulmonary vascular resistance ðPVRÞ ¼ 80 ¼ dynes sec cm5
CO
where 80 is the factor to convert into metric units (dynes sec cm5)
mean aortic pressure mean RAP
System vascular resistance ðSVRÞ ¼ 80 ¼ dynes sec cm5
CO
mean aortic pressure mean RAP
System vascular resistance index ðSVRIÞ ¼ 80 ¼ dynes sec cm5 =m2
CI
Abbreviations: Wt, weight (kg); AO2%, arterial oxygen saturation; VO2%, venous oxygen saturation; Hgb, hemoglobin concentration (g/dL);
BSA, body surface area (m2); PAP, pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; RAP, right atrial pressure.
LAP is in most cases substituted by PCWP. The mean pressure pulmonary venous filling pressure and not to changes in
drop from the PA to the LA or the pulmonary capillaries is the pulmonary vascular bed. If the transpulmonary pressure
termed transpulmonary pressure gradient (Fig. 9.8). In a nor- gradient exceeds the normal range it is an indication of
mal pulmonary vascular bed the transpulmonary pressure pathologic changes in the vascular bed across the lungs
gradient is no larger than 12 mmHg (2). In case of increased (Fig. 9.8) (28). For hemodynamic evaluation and testing of
PA pressures a normal or near normal transpulmonary pres- pulmonary vascular reactivity in case of pulmonary hyperten-
sure gradient indicates that this increase is due to increased sion see chapter 10.
SHUNT DETECTION
Patients with clinically suspected intracardiac shunts nowa-
days undergo echocardiographic examination that usually
reveals the location of cardiac shunts and allows their quan-
tification. Even small, hemodynamically insignificant shunts
like small atrial defects (ASD) or minimal shunts through a
patent foramen ovale (PFO) are visualized by contrast echo-
cardiography or color Doppler echocardiography. Accord-
ingly, the main purposes of right heart catheterization in
patients with known intracardiac shunts are their quantifica-
tion, the assessment of relative shunt volumes in bidirectional
shunts, or the evaluation of concomitant pulmonary hyperten-
sion. Nevertheless, unsuspected intracardiac shunts are occa-
sionally found during routine right heart catheterization. A
left-to-right shunt should be suspected, when PA oxygen
saturation is 80%, a right-to-left shunt, when arterial oxygen
saturation is below 90% or approaches 80%, without apparent
Figure 9.8 Transpulmonary pressure gradient. The measurement
underlying heart failure, pulmonic disorder or alveolar hypo-
of PCWP and PAP in a patient with pulmonary hypertension is
depicted. Transpulmonary pressure gradient (TPG) is defined ventilation. In case of suspected right-to-left shunt, alveolar
as mean PAP minus mean PCWP (e.g., 65 9 ¼ 56 mmHg). An hypoventilation should be excluded by making the patient
increased transpulmonary pressure gradient is an indicator of an cough and taking deep breathes. If arterial desaturation per-
increased vasomotor tone or structural obstructive remodeling of the sists, oxygen should be given by face mask to achieve full
pulmonary vascular bed. Abbreviations: PCWP, pulmonary capillary arterial oxygenation. If full arterial oxygenation can not be
wedge pressure; PAP, pulmonary artery pressure; PA, pulmonary achieved by providing oxygen, right-to-left shunt is presumed
artery. to be present.
[2/7/010/20:36:32] [85–97]
Table 9.4 Detection of Left-to-Right Shunt by Oxygen Saturation Step-Up

O2 % saturation
Level of shunt (chambers) step-up Possible causes of step-up
Atrial (superior vena cava/ 7% ASD, sinus venosus defects, partial anomalous pulmonary venous drainage,
inferior vena cava to RA) coronary or other arteriovenous fistula to the RA, ruptured sinus valsalvae,
VSD with tricuspid regurgitation
Ventricular (RA to RV) 5% VSD, primum ASD, fistula to RV, patent ductus arteriosus with pulmonary
regurgitation
Great vessel (RV to pulmonary 5% Patent ductus arteriosus, aorto-pulmonic window, aberrant coronary artery origin
artery)
Abbreviations: RA, right atrium; ASD, atrial septal defect; RV, right ventricle; VSD, ventricular septal defect.
Source: Adapted from Ref. 29.
Table 9.5 Sample Sites for Oxygen Saturation During Diagnostic the PA and samples should be obtained in the various locations
Oxymetry Run after verification of the position of the catheter by fluoroscopy.
1. Left and/or right PA The possible sample sites are listed in Table 9.5. It is advisable to
2. Main PA obtain two samples of the most important sites (e.g., PA) and to
3. Right ventricular outflow tract sample the receiving chambers several times and average the
4. Right ventricular, mid values. The oxymetry run is performed during steady state
5. Right ventricular, apex conditions within a few minutes. In equivocal cases increasing
6. Right atrium, low or near tricuspid valve systemic flow by exercise will increase the likelihood of detection
7. Right atrium, mid
of a small left-to-right shunt (29). Importantly, supplemental
8. Right atrium, high (near junction with superior vena cava)
oxygen should be withheld. If a significant oxygen step-up and
9. Superior vena cava, low (near junction with right atrium)
10. Superior vena cava, high thus a left-to-right shunt is detected, the blood oxygen saturation
11. Inferior vena cava, high (just beneath heart, above hepatic vein) may be used to calculate the shunt magnitude.
12. Inferior vena cava, low (above renal vein, below hepatic vein) However, it is important to understand, that the step-up
13. Pulmonary vein indicates a net left-to-right shunt. A concomitant right-to-left
14. Left atrium shunt may be present. For example, in large atrial septal defects
15. Left ventricle a right-to-left shunt of varying magnitude is almost always
16. Aorta (distal to insertion of ductus) present. In analogy to the step-up of the oxygen saturation of
Abbreviation: PA, pulmonary artery. the left-to-right shunt, a right-to-left shunt is detected by a step-
down of oxygen saturation in the chambers of the left heart. To
detect and to calculate the magnitude of a right-to-left shunt an
oxygen saturation sample of the pulmonary vein, LA, left
Oxymetry Run ventricle (LV), and aorta (below the insertion of the ductus) is
If a left-to-right shunt or a right-to-left shunt is suspected, it obtained.
should be localized and quantified. In the case of a left-to-right
shunt the receiving heart chamber gets an admixture of
arterial blood that will increase its oxygen saturation as
Calculation of Shunt Size
If an intracardiac shunt is present pulmonary flow (Qp) no
compared with the upstream chamber (Table 9.4). The sim-
longer equals systemic flow (Qs). In the presence of a left-to-
plest way to screen for a left-to-right shunt is to search for an
right shunt pulmonary flow is increased by the shunt volume
oxygen saturation step-up over the entire right-sided heart by
and conversely in the presence of a right-to-left shunt pulmo-
sampling the superior vena cava and the PA. If an oxygena-
nary flow relative to systemic flow is decreased by the shunt
tion step-up of more than 8% from the superior vena cava to
volume. The shunt fraction is then the ratio of the pulmonary
the PA is detected, a left-to-right shunt may be present. For
flow to the systemic flow (Qp/Qs).
exact localization (atrial, ventricular, great vessels) of the
oxygen saturation step-up an oximetry run is performed. On
Calculation of Pulmonic Flow (Qp)
the atrial level an oxygen saturation step-up of 7% and on
The pulmonary flow is calculated according to the Fick for-
mula.
oxygen consumption
Pulmonary flow ðQp Þ ¼ ¼ L= min
ðpulmonary vein pulmonary artery oxygen saturationÞ 1:36 Hgb 10
If no pulmonary vein sample has been obtained arterial
oxygen saturation may be used. A prerequisite for the use of an
ventricular and great vessel one of 5% is an indication of a arterial oxygen saturation as substitute for pulmonary vein
left-to-right shunt (29). saturation is the exclusion of a right-to-left shunt and an arterial
The oxymetry run should be performed with a large bore oxygen saturation of 95%. If a right-to-left shunt is present, an
end-hole or side-hole catheter, usually the Swan-Ganz catheter or assumed oxygen saturation of 98% should be used for the
a multipurpose catheter. The saturation run should be started in calculation of pulmonary blood flow (30).
[2/7/010/20:36:32] [85–97]
Calculation of Systemic Flow (Qs) Quantification of Left-to-Right and Right-to-Left Shunts

The systemic flow is calculated according to the following In absence of a right-to-left shunt the left-to-right shunt is
formula: calculated as pulmonic flow (Qp) minus systemic flow (Qs).
oxygen consumption
Systemic flow ðQs Þ ¼ ¼ L= min
ðsystemic arterial mixed venous oxygen saturationÞ 1:36 Hgb 10
The magnitude of the left-to-right shunt can be expressed as the

The mixed venous oxygen saturation refers to the oxygen
ratio of pulmonic to systemic flow (Qp/Qs). The ratio Qp/Qs
saturation of the heart chamber upstream to the chamber
can be calculated by knowing the oxygen saturations alone
receiving the shunt. For shunts on the atrial level the mixed
venous oxygen saturation is therefore the oxygen saturation of (Table 9.6). The Qp/Qs ratio derived from the oxymetry data is
the superior and inferior vena cava. For shunts on the ven- routinely compared with the Qp/Qs ratio obtained from echo-
cardiographic shunt quantification. Shunt ratios between 1.0
tricular level and for shunts on the great vessel level the mixed
venous saturation used for flow calculation is the oxygen and 1.5 indicate a small left-to-right shunt. A Qp/Qs ratio
saturation of the right atrium, respectively, the RV. between 1.5 and 2.0 an intermediate and a ratio greater than
2.0 a large left-to-right shunt. Surgical or percutaneous defect
In adults the mixed venous oxygen saturation is calcu-
closure is recommended in large and most of the intermediate
lated as the sum of three times the superior vena cava oxygen
saturation plus one time the oxygen saturation of the inferior shunts. Of note, shunt size is not equal to defect size. In the case
of increased right heart pressures left-to-right shunt might be
vena cava divided by four (31). (In pediatric cardiology, the
small despite a large atrial septal defect.
saturation difference between superior and inferior vena cava is
ignored and the saturation of the superior vena cava is usually In the presence of a right-to-left shunt or a bidirectional
used as venous oxygen saturation.) The increased weighting of shunt (simultaneous left-to-right and right-to-left shunts) the
magnitude of each shunt can be calculated by the additional
the more desaturated blood of the superior vena cava is due to
the fact that the admixture of the heavily desaturated blood quantification of the hypothetic effective pulmonic and sys-
from the coronary sinus is not measured. The empiric formula temic blood flow. The effective blood flow is the blood flow that
would exist in a pulmonic or systemic vascular system in
has proven to approximate best the mixed venous saturation in
adults at rest (31). During exercise however, oxygen saturation absence of any left-to-right or right-to-left shunt. The effective
of the inferior vena cava weighs more prominently on mixed blood flow (Qeff) is calculated according to the formula:
oxygen consumption
Effective blood flow ðQeff Þ ¼
ðpulmonic vein saturation mixed venous saturationÞ 1:36 Hgb 10
The left-to-right shunt then equals Qp Qeff and con-

venous oxygen saturation. During exercise mixed venous oxy- versely the right-to-left shunt equals Qs Qeff.
gen saturation therefore is computed as the sum of the oxygen Knowing the effective blood flow allows the calculation
saturation of the vena cava superior plus two times the oxygen of the percentage of shunt volumes. The percentage left-to-
saturation of the inferior vena cava divided by three (31). right shunt equals (1 Qp effective/Qp) and the percentage
Table 9.6 Formulas for Shunt Calculation

3 SVCsat þ 1 IVCsat
Mixed venous oxygen saturation ¼ ¼%
4
O2 consumption
Pulmonary flow ðQp Þ ¼ ¼ L=min
ðPVsat PAsat Þ 1:36 Hgb 10
O2 consumption
System flow ðQs Þ ¼ ¼ L=min
ðArtsat Mixed venoussat Þ 1:36 Hgb 10
Artsat Mixed venoussat
Qp =Qs ¼
PVsat PAsat
O2 consumption
Qp effective ðQs effectiveÞ ¼ ¼ L=min
ðPVsat Mixed venoussat Þ 1:36 Hgb 10
Left-to-right shunt ðQleft-right Þ ¼ Qp Qp effective ¼ L=min
Right-to-left shunt ðQright-left Þ ¼ Qs Qs effective ¼ L=min
PAsat Mixed venoussat
% left-to-right shunt ¼ 100 ¼ %
PVsat Mixed venoussat
PVsat Artsat
% right-to-left shunt ¼ 100 ¼ %
PVsat Mixed venoussat
Abbreviations: SVCsat, blood oxygen saturation of the superior vena cava; IVC, inferior vena cava; PV, pulmonary vein; PA, pulmonary artery;
Art, arterial; Hgb, hemoglobin.
[2/7/010/20:36:32] [85–97]
Table 9.7 Anticipated Normal Values

Right atrium A-wave 2–10 mmHg
V-wave 2–10 mmHg
Mean 0–8 mmHg
Right ventricle Systolic 15–30 mmHg
End diastolic 0–8 mmHg
Pulmonary artery Systolic 15–30 mmHg
End diastolic 3–12 mmHg
Mean 10–21 mmHg
Left atrium or Pulmonary capillary wedge pressure A-wave 4–15 mmHg
V-wave 4–15 mmHg
Mean 4–12 mmHg
Transpulmonary pressure gradient 4–12 mmHg
Cardiac output 5–10 L/min
Cardiac index 3–5 L/min/m2
Stroke volume 80–160 mL
Stroke volume index 40–160 mL/m2
Mixed venous oxygen saturation 65–75%
Systemic vascular resistance 770–1500 dynes sec cm5
Pulmonary vascular resistance 20–120 dynes sec cm5
Note: Normal values are expected values in average sized adults at rest.
Source: From Refs. 26 and 32.
right-to-left shunt is computed as (1 Qs effective/Qs). A 10. Cotter G, Cotter OM, Kaluski E. Hemodynamic monitoring in
simplified formula to calculate percentage of shunts is obtained acute heart failure. Crit Care Med 2008; 36:S40–S43.
by factoring out variables such as oxygen consumption from 11. Grossman W. Pressure measurement. In: Baim DS, ed. Grossman’s
these formulas. The percentage of the respective shunt volumes Cardiac Catheterization, Angiography, and Intervention. 7th ed.
Philadelphia: Lippincott Williams & Wilkins, 2006:133–147.
can then be calculated by using the blood oxygen saturations
12. Kellan EA, Cho L. Right heart catheterization. In: Griffin PB, Topol
alone (Table 9.7). EJ, Nair D, et al., eds. Manual of Cadiovascular Medicine. 3rd ed.
Philadelphia: Lippincott Williams & Wilkins, 2009.
13. Hoeper MM, Lee SH, Voswinckel R, et al. Complications of right
REFERENCES heart catheterization procedures in patients with pulmonary
1. Deanfield J, Thaulow E, Warnes C, et al. Management of grown up hypertension in experienced centers. J Am Coll Cardiol 2006;
congenital heart disease. Eur Heart J 2003; 24:1035–1084. 48:2546–2552.
2. Galie N, Hoeper MM, Humbert M, et al. Guidelines for the 14. Boyd KD, Thomas SJ, Gold J, et al. A prospective study of
diagnosis and treatment of pulmonary hypertension: The Task complications of pulmonary artery catheterizations in 500 consec-
Force for the Diagnosis and Treatment of Pulmonary Hyperten- utive patients. Chest 1983; 84:245–249.
sion of the European Society of Cardiology (ESC) and the Euro- 15. Gupta PK, Haft JI. Complete heart block complicating cardiac
pean Respiratory Society (ERS), endorsed by the International catheterization. Chest 1972; 61:185–187.
Society of Heart and Lung Transplantation (ISHLT). Eur Heart J 16. Sprung CL, Elser B, Schein RM, et al. Risk of right bundle-branch
2009; 30:2493–2537. block and complete heart block during pulmonary artery catheter-
3. Connors AF Jr, Speroff T, Dawson NV, et al. The effectiveness of ization. Crit Care Med 1989; 17:1–3.
right heart catheterization in the initial care of critically ill patients. 17. Burns KE, McLaren A. A critical review of thromboembolic
SUPPORT Investigators. JAMA 1996; 276:889–897. complications associated with central venous catheters. Can J
4. Mueller HS, Chatterjee K, Davis KB, et al. ACC expert consensus Anaesth 2008; 55:532–541.
document. Present use of bedside right heart catheterization in 18. Poplausky MR, Rozenblit G, Rundback JH, et al. Swan-Ganz
patients with cardiac disease. American College of Cardiology. catheter-induced pulmonary artery pseudoaneurysm formation:
J Am Coll Cardiol 1998; 32:840–864. three case reports and a review of the literature. Chest 2001;
5. Harvey S, Harrison DA, Singer M, et al. Assessment of the clinical 120:2105–2111.
effectiveness of pulmonary artery catheters in management of 19. Hardy JF, Morissette M, Taillefer J, et al. Pathophysiology of
patients in intensive care (PAC-Man): a randomised controlled rupture of the pulmonary artery by pulmonary artery balloon-
trial. Lancet 2005; 366:472–477. tipped catheters. Anesth Analg 1983; 62:925–930.
6. Binanay C, Califf RM, Hasselblad V, et al. Evaluation study of 20. Lois JF, Takiff H, Schechter MS, et al. Vessel rupture by balloon
congestive heart failure and pulmonary artery catheterization catheters complicating chronic steroid therapy. AJR Am J Roent-
effectiveness: the ESCAPE trial. JAMA 2005; 294:1625–1633. genol 1985; 144:1073–1074.
7. Sharkey, SW. A guide to interpretation of hemodynamic data in 21. Kaiser CA, Hugli RW, Haegeli LM, et al. Selective embolization of
the coronary care unit, p22. Philadelphia: Lippincott-Raven, 1997. a pulmonary artery rupture caused by a Cournand catheter.
8. O’Quin R, Marini JJ. Pulmonary artery occlusion pressure: clinical Catheter Cardiovasc Interv 2004; 61:317–319.
physiology, measurement, and interpretation. Am Rev Respir Dis 22. Mullerworth MH, Angelopoulos P, Couyant MA, et al. Recogni-
1983; 128:319–326. tion and management of catheter-induced pulmonary artery rup-
9. Mukherjee D, St George D, Knight C, et al. Echocardiography and ture. Ann Thorac Surg 1998; 66:1242–1245.
pulmonary function as screening tests for pulmonary arterial 23. Tuman KJ, Carroll GC, Ivankovich AD. Pitfalls in interpretation of
hypertension in systemic sclerosis. Rheumatology (Oxford) 2004; pulmonary artery catheter data. J Cardiothorac Anesth 1989;
43:461–466. 3:625–641.
[2/7/010/20:36:32] [85–97]
24. Grossman W. Blood flow measurement: cardiac output and vascular 29. Antman EM, Marsh JD, Green LH, et al. Blood oxygen measure-
resistance. In: Baim DS, ed. Grossman’s Cardiac Catheterization, ments in the assessment of intracardiac left to right shunts: a
Angiography, and Intervention. Philadelphia: Lippincott, Williams & critical appraisal of methodology. Am J Cardiol 1980; 46:265–271.
Wilkins, 2006:148–162. 30. Grossman W. Shunt detection and quantification. In: Baim DS, ed.
25. Kendrick AH, West J, Papouchado M, et al. Direct Fick cardiac Grossman’s Cardiac Catheterization, Angiography, and Interven-
output: are assumed values of oxygen consumption acceptable? tion. 7th ed. Philadelphia: Lippincott Williams & Wilkins,
Eur Heart J 1988; 9:337–342. 2006:163–172.
26. Spiller P, Webb-Peploe MM. Blood flow. Eur Heart J 1985; 6(suppl 31. Flamm MD, Cohn KE, Hancock EW. Measurement of systemic
C):11–18. cardiac output at rest and exercise in patients with atrial septal
27. Cigarroa RG, Lange RA, Williams RH, et al. Underestimation of defect. Am J Cardiol 1969; 23:258–265.
cardiac output by thermodilution in patients with tricuspid regur- 32. Kovacs G, Berghold A, Scheidl S, et al. Pulmonary arterial pres-
gitation. Am J Med 1989; 86:417–420. sure during rest and exercise in healthy subjects: a systematic
28. Delgado JF, Conde E, Sanchez V, et al. Pulmonary vascular review. Eur Respir J 2009; 34:888–894.
remodeling in pulmonary hypertension due to chronic heart fail-
ure. Eur J Heart Fail 2005; 7:1011–1016.
[21/6/010/15:27:52] [98–117]
10
Pulmonary hypertension—hemodynamic assessment

and response to vasodilators
Myung H. Park and Vallerie V. McLaughlin
INTRODUCTION FUNDAMENTALS
The first observation of patients with pulmonary hypertension Pathobiology of Pulmonary Arterial Hypertension
was described by a German physician Dr. Ernst von Romberg The pulmonary vasculature is a low-pressure system with
as “sclerosis of the pulmonary arteries” from autopsy findings normal systolic pulmonary artery pressure (SPAP) range of 15
(1). The term primary pulmonary hypertension (PPH) was used to 30 mmHg and mean pulmonary artery pressure (mPAP) of 9
by Dresdale and colleagues in 1951, describing a hypertensive to 18 mmHg, essentially functioning at less than one tenth the
vasculopathy of pulmonary vessels of unknown cause (2). Paul resistance to flow observed in the systemic vascular bed, in part
Wood contributed to understanding of possible etiology of this because of the large cross-sectional area of the pulmonary
disease by observing that a reduction in pulmonary artery circulation (10).
pressure was seen in response to intravenous (IV) administra- The current definition of pulmonary arterial hypertension
tion of acetylcholine in patients with pulmonary hypertension (PAH) from the 4th World Symposium is mPAP > 25 mmHg
secondary to mitral stenosis, eliciting a proposal that a “vaso- and pulmonary capillary wedge pressure (PCWP) 15 mmHg
constrictive factor” may be a cause (3). (11). PAH is characterized by structural changes in the pulmo-
However, it was not until an outbreak of aminorex- nary vascular bed resulting in pulmonary arterial obstruction
induced pulmonary hypertension in 1960s in Europe which due to vascular proliferation and remodeling. This leads to
prompted the World Health Organization (WHO) in assem- progressive increase in pulmonary artery pressure (PAP) and
bling a group of experts to determine the current state of pulmonary vascular resistance (PVR) resulting in right ventric-
knowledge of PPH (4). The National Heart, Lung and Blood ular failure and death. The predominant cause of increased
Institute (NHLBI) created a National Registry of Patients with PVR is the loss of vascular luminal cross-sectional area due to
PPH from 1981 to 1987, enrolling 187 patients from 32 clinical pulmonary vascular remodeling. This process involves all
centers. This registry has had a monumental impact in eluci- layers of the vessel wall and is characterized by intimal hyper-
dating clinical, epidemiological, and pathophysiological infor- plasia, medial hypertrophy, adventitial proliferation, and in situ
mation, which has paved the way for subsequent research. The thrombosis.
data from the registry has revealed that PPH occurs more The process by which pulmonary vasculopathy is initi-
frequently in women than men (1.7:1) with a mean age of ated results from interaction of a predisposing state and one or
diagnosis of 36 years and when left untreated, is a progressive more inciting stimuli, a concept known as “multiple-hit
disease with high mortality with a median survival of 2.8 years hypothesis” (12,13). Two or more “hits” is thought to consist
(5,6). In addition, this study brought to attention that a signif- of a genetic abnormality or substrate that renders an individual
icant delay was reported in making the diagnosis from onset of susceptible. The second hit may be either a systemic disorder
symptoms (2.5 years), a factor which has prompted efforts in [i.e., connective tissue disease (CTD), human immunodefi-
increasing awareness of pulmonary hypertension. ciency virus (HIV)], an environmental trigger (i.e., hypoxic
The 2nd WHO meeting was held in Evian, France in 1998, state, ingestion of an anorexigen), or additional genetic con-
commemorating the 25th anniversary of the first meeting in ditions (i.e., mutation, polymorphism). Once a combination of
Geneva. The experts developed a classification categorizing factors affect a susceptible individual, various mechanisms are
pulmonary hypertension into five groups based on different activated which result in vasoconstriction, cellular proliferation
etiologies. However, the most comprehensive changes were and prothrombotic state leading to PAH.
made during the 3rd World Symposium in Venice, Italy, held
in 2003. This meeting was heralded by tremendous advances in
the field of molecular and genetics sciences, as well as devel- Molecular and Cellular Mechanisms
opment of effective therapies, which changed the understand- Prostacyclin and Thromboxane A2
ing and practice of pulmonary hypertension. The 2003 Venice The two prostanoids, prostacyclin (PGI2) and thromboxane A2,
Classification of Pulmonary Hypertension replaced the term are main metabolites of arachidonic acid. PGI2, produced by the
PPH with idiopathic pulmonary arterial hypertension (IPAH) action of PGI2 synthase, is a potent vasodilator and a strong
along with modifications of the five categories previously inhibitor of platelet aggregation and smooth muscle cell prolif-
established (7). In 2008, the 4th World PH Symposium took eration. Thromboxane A2 is a potent vasoconstrictor and pro-
place in Dana Point, California, where current research and motes platelet activation. In PAH, PGI2 synthase activity and
clinical trials were evaluated resulting in an updated classifi- PGI2 levels are reduced whereas thromboxane level is
cation system and treatment guideline (see “Clinical Aspects”) increased, thereby resulting in vasoconstriction, ceullular pro-
(8,9). liferation and thrombosis (Fig. 10.1) (14–16).
[21/6/010/15:27:52] [98–117]
PULMONARY HYPERTENSION—HEMODYNAMIC ASSESSMENT AND RESPONSE TO VASODILATORS 99
Figure 10.1 Three major mechanistic pathways are known to be perturbed in patients with PAH. (i ) The NO pathway: NO is created in
endothelial cells by type III NO synthase (eNOS), which in turn induces guanylate cyclase (GC) to convert guanosine triphosphate (GTP) to
cyclic guanosine monophosphate, a second messenger that constitutively maintains PASMC relaxation and inhibition of PASMC proliferation.
(ii ) The ET pathway: Big-ET (or pro-ET) is converted in endothelial cells to ET-1 (21 amino acids) by endothelin-converting enzyme (ECE).
ET-1 binds to PASMC ETA and ETB receptors, which ultimately leads to PASMC contraction, proliferation, and hypertrophy. ET-1 also binds
to endothelial cell ETB receptors. (iii ) The prostacyclin pathway: The production of PGI2 is catalyzed by prostacyclin synthase (PS) in
endothelial cells. In PASMCs, PGI2 stimulates adenylate cyclase (AC) thus increasing production from ATP of cAMP, another second
messenger that maintains PASMC relaxation and inhibition of PASMC proliferation. Importanly, the pathways interact as illustrated,
modulating the effect of any single pathway. They also are impacted by transmitters and stimuli that act at cell membrane receptors (Rec).
Examples of these include but are not limited to thrombin, bradykinin, arginine vasopressin (AVP), vessel wall shear stress, angiotensin II (Ang II),
cytokines, and reactive oxygen species (ROS). In addition, the effect of a transmitter depends on its specific site of action (such as PASMC ETA or
ETB receptors vs endothelial cell ETB receptor). The large white arrows depict aberrations in observed in these pathways among patients with
PAH. The PDE5-inh, dual and selective ETRA, and prostanoids are agents that have reported clinically beneficial effects in patients with PAH.
PDE5-Inh indicates phosphodiesterase-5 inhibitor, for example, sildenafil; ETRA, endothelin receptor antagonist, for example, bosentan (dual),
ambrosentan, and sitaxsentan (receptor A selective). Prostanoids, for example, epoprostenol, treprostinil, and iloprost, supplement exogene-
nously deficient levels of PGI2. The dashed lines in octagonal figure signify an inhibitory effect of depicted agents. Dotted arrows depict pathways
with known and unknown intervening steps that are not shown. Abbreviations: PAH, pulmonary arterial hypertension; NO, nitric oxide; PASMC,
pulmonary artery smooth muscle cell; ET, endothelin; PGI2, prostacyclin. Source: From Ref. 16.
Endothelin-1 Nitric Oxide Pathway

Endothelin (ET)-1 is a 21-amino peptide that is produced by NO, produced from arginine by NO synthase in endothelial
endothelium-converting enzymes from big endothelium. ET-1 cells, is a potent selective pulmonary vasodilator. It exerts its
is a potent vasoconstrictor and smooth muscle mitogen and it effects through its second messenger, cyclic guanosine mono-
exerts its effects through two receptors, ETA (located on smooth phosphate (cGMP), which is degraded by phosphodiesterase-5
muscle cells) and ETB receptors (located on vascular endothelial (PDE-5). Patients with PAH have decreased NO synthase activ-
cells and smooth muscle cells) (17,18). Activation of the ETA ity, thus promoting vasoconstriction and cellular proliferation
and ETB receptors on smooth muscle cells induces vasocon- (21). PDE-5 inhibitors (PDE5-Inhs) act by selectively blocking
striction and cellular proliferation and hypertrophy, whereas this enzyme, thus promoting the accumulation of intracellular
stimulation of ETB receptors on endothelial cells results in cGMP and enhancing NO mediated effects (Fig. 10.1).
production of vasodilators [nitric oxide (NO) and PGI2]. ETB
receptors are also involved in the clearance of ET-1 from the Serotonin
circulatory system (18). In PAH patients, plasma levels of ET-1 Serotonin (5-hydroxytryptamine) is a vasoconstrictor that pro-
is increased and its level has been shown to be inversely motes smooth muscle cell hypertrophy and hyperplasia (22).
proportional to the magnitude of the pulmonary blood flow Elevated plasma serotonin and reduced content of serotonin in
and cardiac output (Fig. 10.1) (19,20). platelets have been reported in IAPH and PAH associated with
[21/6/010/15:27:52] [98–117]
ingestion of dexfenfluramine, which increases the release of favorable survival of PAH associated with congenital heart dis-
serotonin from platelets and inhibits its reuptake (23,24). Fur- ease (CHD), polymorphisms in genes related to rennin-angioten-
thermore, mutations in the serotonin transporter (5-HTT) and sin-aldosterone system, and differences in the degree of ischemia
its receptor 5-HT2B have been described in PAH patients (25). and apoptosis (35,38,39). Resurgence of interest in determining
However, it is not certain if elevated serotonin levels are mechanisms of RV failure and more effective method of imaging
implicated in PAH since selective serotonin reuptake inhibitors the RV are currently underway. As for now, one fact remains
(SSRIs) are not associated with an increased incidence of pul- clear which is that RV function is the single most important
monary hypertension and may be even protective against determinant of survival in patients with PAH.
hypoxic PH (26).
Additional Mechanisms INDICATIONS FOR RIGHT HEART

Inhibition of voltage-dependent potassium channels (Kv) have CATHETERIZATION IN PULMONARY ARTERIAL
been linked to factors which promote PAH, such as hypoxia HYPERTENSION
and fenfluamine derivatives (27,28). Abnormalities of the coag- Right heart catheterization is necessary to establish diagnosis,
ulant cascade including increased levels of von Willebrand perform acute vasodilator testing, assess prognosis and guide
factor, plasminogen activator inhibitor-1, and plasma fibrino- therapy in PAH. While echocardiography is the single most
peptide have been reported in PAH patients (29). Furthermore, important screening tool in PH, it lacks the accuracy to make
inflammatory factors such as proinflammatory cytokines and the diagnosis in PAH. Furthermore, the definition of PAH is
autoantibodies have been implicated in PAH (30). based on hemodynamically derived set of parameters: mPAP >
25 mmHg and PCWP 15 mmHg (11).
Genetic Substrates
Molecular genetic studies have identified mutations in a recep- Establishing Diagnosis in Pulmonary Arterial
tor in the transforming growth factor b (TGFb) receptor path- Hypertension
way, named bone morphogenetic protein receptor 2 (BMPR2), Measuring Pulmonary Artery Pressure: Limitations
in certain patients identified with heritable pulmonary arterial of Echocardiography
hypertension (HPAH) (31,32). The mutation in the BMPR2 The method most widely used to screen for PH is measuring the
receptor protein results in aberrations of signal transduction peak systolic velocity of the tricuspid regurgitation (TR) jet with
in the pulmonary smooth muscle cell which leads to cellular continuous-wave spectral Doppler. An estimation of right ven-
proliferation. Less common mutations associated with PAH tricular systolic pressure (RVSP) is generated by adding an
occur in activin like kinase (Alk1), another TGFb receptor assessment of right atrial pressure (RAP) to the gradient using
implicated in patients with hereditary hemorrhagic telangiec- the modified Bernoulli equation: RVSP ¼ 4v2 þ RAP, in which v
tasia and PAH (33). is the velocity of the tricuspid jet in meters per second (10,11,40).
Several factors influence the accuracy of this measure-
The Right Ventricle in Pulmonary Hypertension ment as estimation of SPAP. First, this pressure estimation is
Although it is the pulmonary arterial vasculature where the the RVSP rather than PASP, which is a valid assumption in the
pathological processes take place in PAH, the factor which absence of obstruction to right ventricular outflow (pulmonic
determines symptom and survival rests on the ability of right valve stenosis or outflow tract obstruction). Second factor is the
ventricle (RV) to function under the increased pressure and accuracy of RAP estimation, which can greatly influence the
resistance. The RV is a thin walled, compliant, crescent-shaped RVSP value. Some centers use an arbitrarily fixed value for
structure, formed by the RV free wall [connected to the left RAP, while others employ a clinically estimated value derived
ventricle (LV) by the anterior and posterior septum] and the from the jugular venous pulse (40,41). Another commonly used
interventricular septum. Because of the low resistance of the method is to make an estimation on the basis of the degree of
pulmonary vasculature, the compliant RV is able to pump inferior vena cava collapse during spontaneous respiration.
the same stoke volume as the LV with much less work (34,35). One study suggested that 50% or <50% collapse reflect RAP
The RV must be able to adapt to the increased afterload for values of <10 mmHg or 10 mmHg, respectively (42).
survival in PAH. The initial adaptive response is usually RV Inability to obtain TR jets is another factor and studies
hypertrophy, which has been seen within 96 hours of inducing have demonstrated that the Doppler profile was insufficient to
pulmonary hypertension in animal models (36). RV hypertrophy measure RV to right atrium (RA) pressure gradients in 10% to
can be followed by contractile dysfunction and/or RV dilatation 70% of patients referred for PH evaluation, mainly because of
for further compensatory maneuver to maintain cardiac output poor acoustic windows (43–45). Patients with advanced lung
by increase in preload to offset decrease in fractional shortening. disease are particularly challenging in this regard. Furthermore,
Continued remodeling of the RV soon causes alterations in RV age and weight also affect SPAP in normal individuals. In a large-
shape from crescent to concentric and in flattening out the scale study of 3790 subjects from 1 to 89 years of age, a SPAP >
septum. Because of interventricular dependence, these changes 40 mmHg was found in 6% of those >50 years old and 5% of those
cause LV diastolic dysfunction and decrease in LV end diastolic with a BMI > 30 kg/m2 (46). The level of physical training has
volume, resulting in further decline in stroke volume and dete- also been shown to affect SPAP. Comparing a group of highly
rioration of end organ function (37). trained athletes versus normal males, SPAP was higher among
However, the development of RV failure due to PAH is the trained individuals both at rest and with exercise, largely
quite variable and the reasons why some RV can compensate because of increases in stroke volume affecting pulmonary artery
maintaining adequate cardiac output for prolonged periods of pressures (47).
time while others immediately decompensate remains unclear. While some studies comparing Doppler-derived SPAP
Several mechanisms have been proposed including retention with catheterization have reported good correlation, others
of the “fetal” genotype believed to be contributory factor for have demonstrated substantial discrepancy between the
[21/6/010/15:27:52] [98–117]
techniques (44,48–54). In patients with severe PH, Doppler- Accurate measurement of left heart filling pressure is
derived SPAP has been shown to commonly underestimate critical for correct diagnosis of PAH. Definition of PAH
pressures (50). With advanced lung disease and PH, SPAP requires both elevation of mPAP (>25 mmHg) and normal
measurements frequently overestimated true PAPs leading to PCWP (or LVEDP of 15 mmHg). The difference between these
over diagnosis (51–53). A similar lack of adequate correlation two measurements calculates the transpulmonary gradient
was reported in patients with PH associated with systemic (TPG ¼ mPAP PCWP). Elevated PCWP is characteristic of
sclerosis (54). PH in the setting of chronically elevated left-sided cardiac
filling pressure, termed pulmonary venous hypertension
Essential Components of a Complete (PVH), and is classified as WHO group 2 PH (8). PVH usually
Hemodynamic Assessment in Pulmonary results from systolic and/or diastolic cardiac dysfunction or
Arterial Hypertension valvular disease. Thus therapeutic decisions can be signifi-
Although the basic principles of right heart catheterization cantly different on the basis of the left-sided filling pressure
were discussed in chapter 9, some comments specific to PAH measurement. PAH is characterized by elevated PAPs, normal
are appropriate. The most common abnormality in RAP in PCWP and elevated TPG whereas in PVH, PAPs are elevated
PAH is due to TR, which can produce attenuated x descent, but TPG is normal because of elevated PCWP.
prominent c-v wave and a deep and rapid y descent (Fig. 10.2) Careful attention to wave forms and timing of measure-
(55,56). In severe TR, ventricularization of RAP may occur ment is essential for accuracy. The most common mistake is
where RAP is nearly indistinguishable from the RV pressure “underwedging,” which occurs with incomplete advancement
contour (Fig. 10.2). In PAH with RV hypertrophy and volume of the PA catheter resulting in a hybrid tracing of PAP and
overload, a prominent a wave may appear on the ventricular PCWP. This usually results in a falsely elevated PCWP, leading
waveform at end-diastole indicative of RA contracting against a to misdiagnosing a patient as PVH (Fig. 10.4). If an operator is
noncompliant RV. A careful assessment of RAP is imperative suspicious that the PCWP being measured is greater than
since RAP carries a significant prognostic importance in PAH. expected on the basis of clinical assessment, an oxygen saturation
In advanced PAH, the PAPs can be elevated to various degrees measurement can be done from the distal port with the catheter
and can reach systemic levels (Fig. 10.3). The PA diastolic in the wedge position. Its measurement should be equal or close
pressure does not correlate well with the mean PCWP in the to the systemic arterial oxygen saturation (usually >90%) done
presence of pulmonary vascular disease. by pulse oximetry. If it is markedly lower, the catheter is most
Figure 10.2 (A) Right atrial waveform from a patient with secondary tricuspid regurgitation from associated severe left-sided heart failure
and right-sided heart failure. Attenuation of the x descent is present, leading to prominent c-v wave. (B) These tracings are from a paitent with
severe tricuspid regurgitation. The right atrial waveform shows ventricularization (left). Compare with the right ventricle (RV) waveform from
same patient (right ). Abbreviation: RA, right atrium. Source: From Ref. 56.
[21/6/010/15:27:52] [98–117]
challenging. One helpful maneuver is deflating the balloon,

allowing the catheter to migrate distally, and carefully reinflating
the balloon following the pressure tracings closely. Usually with
this approach, optimal placement is obtained with balloon
partially inflated. An intraluminal guide wire can also aid in
advancing the catheter to a more distal position. All these
maneuvers should be performed very cautiously and under
direct fluoroscopic visualization since patients with PAH are at
increased risk of PA rupture, a potentially fatal event.
Characteristics of a physiologically reliable PCWP and
steps that can aid in obtaining the measurement include the
following (55–57): (i) A distinct a and v waves should be
present. Exception noted in atrial fibrillation where a wave
will be absent. (ii) Waiting for steady state in PCWP tracing to
occur (not immediately after the balloon is inflated) and record
at end-expiratory phase. (iii) A distinct immediate rise in pres-
sure when balloon is deflated out of the wedge position. (iv)
Catheter tip should be stable in PA when viewed under fluo-
roscopy with the balloon inflated (not moving back and forth).
(v) An oxygen saturation measured in PCWP > 90%.
(vi) Multiple measurements of PCWP with similar results. If
these maneuvers fail to obtain a reliable PCWP, a left heart
catheterization should be performed to measure LVEDP.
Although less common, the catheter can also be “over-
wedged” with excessive inflation of the balloon relative to the
size of the vessel. This should be avoided not only because of
inaccurate pressure measurement but also because of increase in
risk of vessel rupture. In bedside catheter measurements, the
potential for PA catheter migration also need to be kept in mind.
The balloon should be slowly inflated at every measurement with
close monitoring of the pressure tracings, with inflation stopped
Figure 10.3 Pulmonary artery pressure and pulmonary capillary
wedge tracing from a patient with severe pulmonary arterial hyper- when a PCWP tracing is obtained.
tension. PA pressure 102/34 mmHg with mean of 63 mmHg and The presence of a “large” v wave can also lead to inac-
PCWP 7 mmHg. Each line represents 10 mmHg. curate reading of PCWP. The v wave is a normal finding on the
wedge tracing and normally higher than the a wave so as to
what measurement constitutes a large v wave is subjective.
Common causes of a large v wave include mitral regurgitation
likely underwedged. In patients with significant elevation of PAP (MR), though the height of the v wave is neither sensitive nor
and/or PA that is dilated, placing the catheter in the correct specific indicator of the degree of MR (58,59). Other causes
anatomic position for optimal PCWP measurement can be include any situations which increase volume or flow into a
Figure 10.4 Differences between correct pulmonary capillary wedge pressure and underwedged tracings.
[21/6/010/15:27:52] [98–117]
noncompliant left atrium, such as ventricular septal defect, Acute Vasodilator Testing
mitral stenosis, cardiomyopathy of any etiology, or postopera- The purpose of evaluating PAH patients with a short-acting
tive surgical conditions. vasodilator is to determine the degree in which the pulmonary
Accurate cardiac output measurement is critical in calcu- vasoconstriction is contributing to the elevated PAPs. Vaso-
lating PVR and assessing prognosis in PAH. The total pulmo- dilator responsiveness identifies patients with a better progno-
nary resistance (TPR) calculates the relationship between the sis and those who are more likely to have a sustained beneficial
mPAP and CO: TPR ¼ mPAP 80/CO; the normal TPR is 100 response to oral calcium channel blockers (CCBs).
to 300 dynes-sec/cm5. The PVR measures the resistance to flow IV epoprostenol and IV adenosine have both been studied
imposed by pulmonary vasculature without the influence of as acute vasodilators. Both are short-acting, potent vasodilators
the left-sided filling pressure: PVR ¼ (mPAP PCWP) 80/ and investigators have reported different degrees of responsive-
CO or PVR ¼ TPG 80/CO; the normal PVR is 20 to 130 ness depending on the criteria used (63–65). However, because
dynes-sec/cm5. Cardiac output assessment is a critical measure both agents have the potential to cause systemic hypotension and
of RV performance in the presence of elevated PAPs and low side effects, using inhaled NO emerged as the vasodilator of choice
cardiac output is a marker of poor prognosis (see Assessment of because of its pulmonary selectivity, short half-life, and lack of
Prognosis) (6). systemic side effects (Table 10.1) (66). However, it is expensive and
Chronic left-to-right intracardiac shunting can result in requires trained respiratory personnel to administer.
PAH. Echocardiogram with agitated saline contrast can detect The definition of what constitutes acute vasodilator “res-
right-to-left shunt but can fail to detect left-to-right shunts. ponder” has undergone changes over the years. The current
Multiple measurements of oxygen saturations from superior consensus definition is a fall in mPAP of at least 10 mmHg to
and inferior vena cava, RA and PA can detect and quantify 40 mmHg, with an increased or unchanged cardiac output
shunts (see chap. 9). If the shunt has reversed, the typical (67,68). If a patient meets this acute criteria, they should be
“step-up” may not be present. A detailed oxygen saturation treated with oral CCBs., however, need to be followed closely
study is a crucial part of right heart catheterization in a patient for a clinical response. Those who improve to functional class
with clinical or echocardiographic suspicion of intracardiac (FC) I or II without the need for additional therapy are likely to
shunting. do well. However, this response is rare, approximately 6.8% of
IPAH patients in a large French series (67). Patients who do not
meet the definition of an acute response should not be treated
Assessment of Prognosis with CCBs.
Since PAH is a disease manifested by an increase in afterload of Acute vasodilator response is very rare in patients with
pulmonary arteries leading to progressive right ventricular associated forms of PAH. Patients with advanced disease such
dysfunction and failure, hemodynamic markers are considered as FC IV symptoms, overt right heart failure, or hemodynamic
to be the gold standard for indicating prognosis. This was first markers of advanced process (high RAP and/or reduced CO,
demonstrated in the NIH Registry where the investigators systemic hypotension) should not undergo acute vasodilator
concluded, “Mortality was most closely associated with right testing since these patients need prompt treatment with PAH-
ventricular hemodynamic function and can be characterized by approved therapies and are not appropriate candidates for
means of an equation using three variables: mean pulmonary CCBs. The development of acute pulmonary edema during
artery pressure, mean right atrial pressure, and cardiac index” vasodilator testing should raise the suspicion of veno-occlusive
(6). Specifically, RAP 20 mmHg, mPAP 85 mmHg, and disease or pulmonary capillary hemangiomatosis, in which
cardiac index (CI) < 2 L/min/m2 were associated with an therapy with pulmonary vasodilator is contraindicated (69).
increased risk of death. The data obtained were the basis of
formulating the regression equation to calculate survival on the
basis of hemodynamics, which was validated in a prospective Risks Associated with RHC in PAH Patients
study (60). Subsequent studies have corroborated the impor- Although RHC is necessary for correct diagnosis of PAH,
tance of elevated RAP and low CO as determinants of poor concerns regarding risks in this population have been raised.
outcome (61). The relevance of mPAP on prognosis has been A recent multicenter study, which included 15 PAH centers
variable. In the retrospective study among patients treated with over five-year period with >7000 procedures, evaluated the
epoprostenol, patients with lower mPAP correlated with poor safety and risks of this procedure (70). The overall incidence of
outcome which may indicate that mPAP per se is not a reliable serious adverse events was 1.1%. The most frequent complica-
surrogate for RV function but needs to be assessed as part of tions were related to venous access; others included arrhythmia
PVR (mPAP/CO) (62). and hypotension due to vagal reactions or pulmonary
Table 10.1 Agents Used in Acute Vasodilator Testing in Patients with Pulmonary Arterial Hypertension
Nitric oxide Epoprostenol Adenosine
Route of administration Inhaled IV IV
Dose range 10–80 ppm 2–10 ng/kg/min 50–250 mcg/kg/min
Dosing increments None to variable titration 1–2 ng/kg/min every 10–15 min 50 mcg/kg/min every 2 min
(10–80 ppm for 5–10 min)
Side effects Increased left-sided filling Hypotension, headache, flushing, Chest tightness, dyspnea,
pressure in susceptible nausea, lightheadedness atrioventricular block, hypotension
patients
[21/6/010/15:27:52] [98–117]
vasoreactivity testing. Thus the authors concluded that when Table 10.2 Risk Factors Favoring Diagnosis of Diastolic Heart Failure
performed in experienced centers, RHC in PAH patients are
Clinical features
safe and associated with low morbidity rates.
Age >65 yr
The following maneuvers can enhance safety of the pro-
Elevated systolic blood pressure
cedure. When accessing from an internal jugular approach, use Elevated pulse pressure
of an ultrasound device to visualize the size and depth of the Obesity
vein greatly assists in gaining access safely. Since PAH patients Hypertension
often have dilated right-sided chambers which can make Coronary artery disease
maneuvering the catheter difficult, especially under high pres- Diabetes mellitus
sure systems and under significant tricuspid valvular regurgi- Atrial fibrillation
tation, performing the procedure under fluoroscopy reduces Echocardiography
the risks of catheter “coiling” and inducing arrhythmia. Direct Left atrial enlargement
Concentric remodeling (relative wall thickness > 0.45)
visualization also assists in placing the catheter in the safe and
Left ventricular hypertrophy
optimal “wedge” position to avoid PA rupture, “overwedging”
Elevated left ventricular filling pressures (grade II to IV diastolic
and migration of the catheter. Fluoroscopy is also necessary in dysfunction)
patients with intracardiac devices. Furthermore, having periph- Interim evaluation (after echocardiography)
eral IV access in patients prior to starting the procedure is Symptomatic response to diuretic drugs
recommended to promptly deliver treatment in the event of Exaggerated increase in systolic blood pressure with exercise
vagal episodes, which can lead to significant clinical deteriora- Rereview of chest radiograph consistent with heart failure
tion in PAH patients. Source: From Ref. 76.
Evaluating PH with Left-Sided Heart Disease be normal after treatment with diuretics. Some investiga-
Diastolic Dysfunction and Pulmonary Hypertension tors have advocated fluid challenge or exercise to assess
Diastolic heart failure (DHF) refers to clinical syndrome in response as a measure of the LV compliance. Although
which patients present with heart failure symptoms with there are no definite standardizations, the recently pub-
preserved left ventricular systolic function. Epidemiological lished ACCF/AHA Expert Consensus Document on Pul-
studies have shown high prevalence of DHF (*40–70%) monary Hypertension and reports from the 4th World
among symptomatic patients and the risk factors have been Symposium on Pulmonary Hypertension outline consen-
well elucidated (age >65, hypertension, elevated pulse pres- sus-based recommendations for evaluation of patients
sure, obesity, coronary artery disease, diabetes mellitus, atrial presenting with both syndromes (Fig. 10.5) (68,76).
fibrillation) (71,72). The predominant underlying structural 2. PCWP is elevated (>15 mmHg) and PVR is <3 WU and
abnormalities in DHF are concentric remodeling and hypertro- TPG is normal, the patient has diastolic dysfunction and
phy of the LV caused by chronic pressure overload, usually due therapy should be aimed at optimizing volume status,
to systemic hypertension. These alterations produce abnormal- heart rate and systemic blood pressure.
ities in both relaxation and filling, which can be a precursor to 3. If the PCWP and the PVR are both elevated (the TPG can be
LV systolic dysfunction or be the main structural abnormality normal or elevated), careful evaluation and intervention
producing symptoms and signs of heart failure (73,74). need to be made to determine if the elevated PVR is passive
Patients presenting with diastolic dysfunction and PH is (because of elevated filling pressure and thus responsive to
a common clinical dilemma and can be very challenging to diuretics and/or systemic vasodilator) or fixed (remain
distinguish from PAH. Up to 70% of patients with LV diastolic elevated despite normalizing PCWP and systemic blood
dysfunction may develop PH, the presence of which is asso- pressure). If the PCWP and PVR both decrease (TPG nor-
ciated with a poor prognosis (75). The presentations are similar mal) with optimal heart failure therapy, then patients need
to PAH and include dyspnea and/or signs and symptoms of to be treated aggressively with these regimen. If the PCWP
heart failure. Echocardiographic findings suggestive of LV normalizes but PVR remains elevated (elevated TPG), this
diastolic dysfunction include left atrial enlargement, LV hyper- may be indicative of pulmonary arteriopathy being the
trophy and elevated LV filling pressure (grade II to IV diastolic dominant disorder with structural changes in pulmonary
dysfunction) (Table 10.2) (68,76). vasculature along with diastolic dysfunction.
At this juncture, it is critical to perform RHC to measure
the left-sided filling pressure and calculate the TPG and PVR. It No PAH-specific therapies have been systematically
needs to be emphasized that attention must be paid to the quality studied for PH associated with diastolic dysfunction. In
of the PCWP tracing for correct diagnosis to be made. Misinter- patients with chronic heart failure, treatment with epoprostenol
pretation of either “underwedged” or hybrid tracing as true and endothelin receptor antagonists (ERAs) have failed to show
PCWP (thereby misdiagnosing as diastolic dysfunction because beneficial effects, though these trials did not specifically target
of falsely elevated PCWP) or recorded measurements from patients with heart failure and PH (77–79). The PDE5-Inh
improper placement of the catheter can lead to wrong diagnosis. sildenafil has shown improvement in LV systolic and diastolic
The possible results obtained fall into one of three cate- function as well as systemic vasoreactivity in animal models of
gories. heart failure (80,81). Recent small, short-term studies evaluat-
ing patients with chronic systolic heart failure and PH using
1. PCWP is normal (<15 mmHg) and TPG and PVR is sildenafil have demonstrated improvement in exercise capacity
elevated [3 Wood units (WU)], the patient has PAH and quality of life (82,83). However, data from a well-designed
and treatment need to be considered after full evaluation. trial studying long-term benefits is necessary before any rec-
If the patient has clinical risk factors and/or echo findings ommendations can be made in regards to use of sildneafil in
suggestive of diastolic dysfunction, PCWP or LVEDP can patients with heart failure and PH.
[21/6/010/15:27:52] [98–117]
Figure 10.5 Diagnostic approach to distinguish between PAH and PH caused by diastolic left heart disease. Abbreviations: DHF, diastolic
heart failure; Dx, diagnosis; EF, ejection fraction; HF, heart failure; NTG, nitroglycerine; OMT, optimal medical therapy; PAH, pulmonary
arterial hypertension; PCWP, pulmonary capillary wedge pressure; PH, pulmonary hypertension; PVR, pulmonary vascular resistance; RCT,
randomized controlled trial; RHC, right heart catheterization; WU, Wood units. Source: From Ref. 76.
Systolic Heart Failure and Pulmonary Hypertension: Evaluation for after HTx is a continuous positive one with the risk of death
Cardiac Transplantation rising with increase in PASP and PVR for early and late trans-
PH secondary to LV systolic dysfunction is a common compli- plant outcomes (88). In determining hemodynamics, the values
cation in patients presenting with advanced HF and RV dys- reported to define “reversible” from “fixed” PH are variable
function (75,84). This process is mainly due to chronic increase and is center dependent.
in left-sided filling pressure resulting in perturbations in vas-
1. PASP > 50 mmHg despite optimal vasodilation has been
cular mediators resulting in increase in vascular tone and
reported to be a relative contraindication to HTx (89).
structural remodeling. Typically the duration and severity of
2. Different PVR values have been reported to be associated
heart failure is an important determinant that governs the
with adverse outcome. PVR > 4 WU has been reported to
degree of these changes, with abnormalities in vascular tone
be an independent predictor of early post transplant mor-
being the early phase that is manageable with vasodilator
tality (90). PVR < 5 WU at rest or < 3 WU with maximal
agents (reversible PH), and structural changes appearing at
vasodilatation is considered favorable.
more advanced stage usually not amenable to pharmacological
3. TPG is viewed by some centers to be a more reliable
maneuvers (fixed PH) (85,86).
marker for pulmonary vascular tone since it does not
Preoperative assessment of PH is a critical part of heart
rely on CO. TPG 16 mmHg has been shown to have
transplant (HTx) evaluation since preoperative PVR is an inde-
an increased risk of postoperative RV failure (84).
pendent risk factor for early mortality after HTx (87). The
degree of pulmonary vascular changes in the recipient is a Agents used to test vasoreactivity differ widely ranging
major factor that determines the RV function post transplant from systemic vasodilators to more selective pulmonary vaso-
and RV dysfunction accounts for both early deaths and post- dilators. One needs to consider various factors which include
operative complications (88). It is imperative to determine if the systemic blood pressure, severity of PH, CO, and clinical sta-
elevation in PVR is reversible and manageable with pharma- bility in determining the optimal agent (91,92). Some investi-
cological therapies to avoid donor heart RV failure from sub- gators have also advocated combining more than one agent to
jecting to acute rise in pulmonary vascular tone of the recipient. target multiple hemodynamic abnormalities to determine
There is no absolute or reliable hemodynamic threshold degree of reversibility. For patients who are determined to
below which RV failure is avoidable or beyond which it is have severe PH despite maximal medical therapy, consider-
certain to occur. The relationship between PH and mortality ation for mechanical unloading with left ventricular assist
[21/6/010/15:27:52] [98–117]
devices (LVAD) is the next step. Mechanical unloading of the PAH has been reported in 6% to 10% of patients with PAH (5). It
LV has been shown to decrease PVR by inducing reverse is characterized by autosomal dominant transmission, incom-
remodeling of pulmonary vascular through alleviating chronic plete penetrance and genetic anticipation, in which family mem-
elevation of left-sided filling pressures and by improving oxy- bers of successive generation who develop PAH manifest at an
genation and CO. These effects are not immediate and optimal earlier age with more aggressive disease course. The mutation in
changes reported to occur between two and six months of BMPR2 loci is the most widely studied and has been identified
support. Successful orthotopic HTx after LVAD placement in not only in patients with familial PAH (50–90%) but also among
patients with fixed PH have been reported (93–95). Both pulsa- 25% of IPAH patients, raising the possibility of spontaneous
tile and continuous flow VAD systems have been used. mutations in some individuals or familial transmission among
members without clinically evidence disease (13,96,97).
Although the incidence of IPAH is rare, PAH has been
CLINICAL ASPECTS identified to occur with increased frequency in the presence of
Classification of Pulmonary Hypertension CTD, HIV, portal hypertension and CHD. Patients with CTD,
The clinical classification of PH from the 4th World Symposium especially the scleroderma spectrum, comprise the largest sub-
at Dana Point has several modifications to the Venice Classifi- group of population affected. Patients with PAH associated
cation (Table 10.3) (8). Some key changes include subclassifica- with CTD have poorer survival than IPAH patients; median
tion of heritable PH and addition of chronic hemolytic anemia survival of 12 months have been reported compared with
and schistosomiasis to the associated PAH category under 2.6 years in IPAH patients (5,98). Furthermore, current thera-
WHO Group 1. pies are less effective in CTD patients compared with IPAH
patients (99).
Pulmonary Arterial Hypertension PAH associated with CHD occurs as a result of high
IPAH is PAH of unknown cause, a diagnosis of exclusion deter- pulmonary blood flow from systemic-to-pulmonary shunts and
mined after a thorough evaluation. IPAH is more common from smaller lesions such as atrial septal defect. Portopulmo-
among young females as reported from the NIH registry (F:M nary hypertension is PAH that occurs in association with liver
1.7:1, mean age 37 years), though the age of affected individuals disease and portal hypertension and is reported in 4% to 15% of
appear to be increasing likely reflecting increased awareness of patient being evaluated for liver transplantation (100,101). Por-
the disease and improved survival with therapy (5,8,9). Heritable tal hypertension results in a high cardiac output state, so in
Table 10.3 Clinical Classification of PH from Dana Point Meeting, 2009

1. Pulmonary arterial hypertension
1.1 Idiopathic PAH
1.2 Heritable
1.2.1. BMPR2
1.2.2. ALK1, endoglin (with or without hereditary hemorrhagic Telangiectasia)
1.2.3 Unknown
1.3 Drug- and toxin-induced
1.4 Associated with
1.4.1. Connective tissue diseases
1.4.2 HIV infection
1.4.3 Portal hypertension
1.4.4 Congenital heart diseases
1.4.5 Schistosomiasis
1.4.6 Chronic hemolytic anemia
1.5 Persistent pulmonary hypertension of the newborn
10 Pulmonary veno-occlusive disease (PVOD) and/or pulmonary capillary hemangiomatosis (PCH)
2. Pulmonary hypertension due to left heart disease
2.1 Systolic dysfunction
2.2 Diastolic dysfunction
2.3 Valvular disease
3. Pulmonary hypertension due to lung diseases and/or hypoxia
3.1 Chronic obstructive pulmonary disease
3.2 Interstitial lung disease
3.3 Other pulmonary diseases with mixed restrictive and obstructive pattern
3.4 Sleep-disordered breathing
3.5 Alveolar hypoventilation disorders
3.6 Chronic exposure to high altitude
3.7 Developmental abnormalities
4. Chronic thromboembolic pulmonary hypertension (CTEPH)
5. PH with unclear multifactorial mechanisms
5.1 Hematological disorders: myeloproliferative disorders splenectomy.
5.2 Systemic disorders, sarcoidosis, pulmonary Langerhans cell histiocytosis, lymphangioleiomyomatosis, neurofibromatosis, vasculitis
5.3 Metabolic disorders: glycogen storage disease, Gaucher disease, thyroid disorders
5.4 Others: tumoral obstruction, fibrosing mediastinitis, chronic renal failure on dialysis.
Abbreviation: PH, pulmonary hypertension.
[21/6/010/15:27:52] [98–117]
general, the cardiac outputs of portopulmonary hypertension natriuretic peptide (BNP) and N-terminal pro-BNP (NT-pro-
patients tend to be higher than other types of PAH. A normal BNP) have been shown to correlate with outcome, though the
cardiac output in a portopulmonary hypertension patient sug- specifics of how to utilize this marker is still under investiga-
gests right ventricular dysfunction. The degree of PAP elevation (110–112). It is recommended that a composite of subjective
tion (mPAP > 35 mmHg) has significant impact on peri- and objective data be used to determine risk of the patient
transplant morbidity and survival (102). Regarding toxic which can be used to choose appropriate therapy and as a
agents, a definite association between ingestion of amphet- guide to determine response to treatment (Table 10.4) (16,68).
amine-derived drugs and PAH has been established, the most
notable ones being appetite suppressants aminorex, fenflur-
amine, and dexfenfluramine (24). All of these agents have been Treatment of PAH
removed from the market after studies demonstrated linkage Prostanoids
between these drugs and PAH. An association between meth- Epoprostenol IV epoprostenol improves FC, exercise
amphetamine use and PAH has been reported recently as well capacity, hemodynamics, and survival in IPAH, which was
(8,103). HIV infection is a risk for PAH with approximately 1 of demonstrated in an open-label, randomized trial of 81 FC III
200 patients being affected (104). Among patients with PAH, and IV IPAH patients comparing IV epoprostenol with con-
survival is the worst for patients with CTD and HIV (61). ventional treatment (106). All eight deaths during the 12-week
trial period occurred among patients who were randomized to
conventional therapy, which resulted in a survival benefit (P ¼
Evaluation of PAH
0.003). IV epoprostenol has also been studied in PAH associ-
Evaluating patients with suspected PH encompasses recogniz-
ated with CTD demonstrating marked improvements in
ing at risk population, screening for PH, identifying the under-
6MWD and hemodynamics but no effect on mortality in a 12-
lying cause or associated disease, and confirming diagnosis and
week, open-label randomized trial (113). Observational studies
assessing prognosis. A diagnostic approach including pivotal
have also reported beneficial effects of IV epoprostenol in
and contingent tests and rationale is shown in Figure 10.6. The
patients with PAH related to HIV, CHD, and portopulmonary
right heart catheterization is required to make the diagnosis of
hypertension (114–116). Two longer-term observational studies
PAH.
have confirmed the chronic benefits of IV epoprostenol in IPAH
patients, specifically improvements in survival compared with
Prognostic Indicators in PAH historical controls, FC, 6MWD, and hemodynamics (62,105).
Prognosis in PAH is related to RV function and indicators IV epoprostenol is a challenging therapy to implement
utilized to assess include WHO FC, exercise capacity, and because of its short half-life (<6 minutes) and the need for
hemodynamics (16). The importance of RAP, mPAP, and CO continuous IV infusion via a tunneled catheter. Each patient
as critical determinants of outcome as initially shown in the must learn the techniques of sterile preparation of the medica-
NIH registry has been discussed in prior section (“Assessment tion, operation of the ambulatory infusion pump, and care of
of Prognosis”). The NIH registry also demonstrated that sur- the central venous catheter. Incidences of sepsis and catheter
vival correlated directly with FC: for patients who were in FC I related infections are not negligible (0.1–0.6 case per patient
or II at presentation, the median survival was almost 6 years year) and can cause significant morbidity (105). Any interrup-
versus 2.5 years for patients in FC III and six months for tion of the drug infusion can be potentially life threatening
patients presenting in FC IV (6). Even on therapy, FC was because of short half-life of epoprostenol and potential for
shown to be an important determinant in two large retrospec- rebound pulmonary hypertension. IV epoprostenol is com-
tive studies among IPAH patients receiving epoprostenol in monly started in the hospital at a dose of 2 ng/kg/min and
that prognosis was worse for patients who were initiated on titrated on the basis of PAH symptoms and side effects. Most
therapy with more advanced symptoms (62,105). Furthermore, experts consider optimal dose of chronic therapy to be between
patients who improved to FC I or II after the initial period 25 and 40 ng/kg/min. Chronic overdose can lead to high
(3–17 months) had a significantly better long-term prognosis cardiac output failure and lead to recurrent symptoms (117).
than those who remained in FC III or IV on IV epoprostenol. Common side effects include headache, jaw pain, diarrhea,
The six-minute walk distance (6MWD) is the most com- nausea, flushing, rash, and musculoskeletal pain. Because of
monly used test to evaluate exercise capacity in PAH. In the the complexity of administering this therapy, epoprostenol use
first PAH trial evaluating treatment with epoprostenol, baseline should be limited to experienced centers (Table 10.5) (118).
6MWD was a powerful predictor of survival (106). Sitbon and Treprostinil Treprostinil is a PGI2 analogue with a half-
colleagues report among patients treated with epoprostenol, life of four hours which was studied as a continuous subcuta-
the 6MWD performed after three months of therapy correlated neous infusion in a 12-week, placebo-controlled, randomized
with long-term survival; specifically, patients who walked trial of 470 patients with FC II, III, or IV PAH (119). There was a
380 m demonstrated a significantly better outcome than the modest but statistically significant median increase of 16 m in
cohorts who did not (62). Used less commonly mainly because 6MWD; the improvement was dose related and patients in the
of patient limitations, cardiopulmonary exercise testing has highest dose quartile reported close to 40 m improvement.
also been studied and one study demonstrated that maximum However, the major hindrance of using subcutaneous trepros-
oxygen consumption of >10.4 mL/kg/min and peak systolic tinil is the pain and erythema at the infusion site, which was
BP > 120 mmHg to be favorable indicators (107). reported by 85% of the patients and limited the dose increases.
Echocardiographic findings correlating with prognosis It is now recognized that site pain is not dose related, and that
include measurements reflecting right-sided cardiac function some patients feel better after proper dose escalation which
(right atrial and right ventricular size, Doppler parameters and helps them to improve their PAH symptoms.
indices of RV function, eccentricity index) and presence of Because of limitation of subcutaneous delivery system,
pericardial effusion (108,109). Biomarkers, specifically brain IV treprostnil was studied in a 12-week open-label trial of
[21/6/010/15:27:52] [98–117]
Figure 10.6 Diagnostic approach to PAH. General guidelines for the evaluation of pulmonary hypertension. Since the suspicion of PH may
arise in various ways, the sequence of tests may vary. However, the diagnosis of PAH requires that certain data support a specific diagnosis.
In addition, the diagnosis of idiopathic pulmonary arterial hypertension is one of excluding all other reasonable possibilities. Pivotal tests are
those that are essential to establishing a diagnosis of any type of PAH either by identification of criteria of associated disease or exclusion of
diagnosis other than IPAH. All pivotal tests are required for a definitive diagnosis and baseline characaterization. An abnormality of one
assessment (such as obstructive pulmonary disease on PFTs), does not preclude that another abnormality (chronic thromboembolic disease
on VQ scan and pulmonary angiogram) is contributing or predominant. Contingent tests are recommended to elucidate or confirm results of
the pivotal tests, and need only be performed in the appropriate clinical context. The combination of pivotal and appropriate contingent tests
contribute to assessment of the differential diagnosis in the right-hand column. It should be recognized that the definitive diagnosis may
require additional specific evaluation not necessarily included in this general guideline. Abbreviations: 6MWT, six-minute walk test; ABGs,
arterial blood gases; ANA, antinuclear antibody serology; CHD, congenital heart disease; CPET, cardiopulmonary exercise test; CT,
computed tomography; CTD, connective tissue disease; CXR, chest X ray; ECG, electrocardiogram; HIV, human immunodeficiency virus
screening; Htn, hypertension; LFT, liver function test; PE, pulmonary embolism; PFT, pulmonary function test; PH, pulmonary hypertension;
RA, rheumatoid arthritis; RAE, right atrial enlargement; RH Cath, right heart catheterization; RVE, right ventricular enlargement; RVSP, right
ventricular systolic pressure; SLE, systemic lupus erythematosis; TEE, transesophageal echocardiography; VHD, valvular heart disease; VQ
scan, ventilation-perfusion scintigram. Source: From Ref. 68.
16 patients (120). It demonstrated improvements in 6MWD treprostinil at the end of the study period was more than twice
(82 m) and hemodynamics. In another open-label trial, 31 FC the dose of epoprostinil at the start of the study. Inhaled
II and III PAH patients on IV epoprostenol were transitioned to treprostinil was recently approved by the FDA and oral tre-
IV treprostinil (121). Twenty-seven patients completed the prostinil is currently undergoing active clinical investigation
transition, and four were transitioned back to epoprostenol. (Table 10.5).
6MWD measurements were maintained among patients who Inhaled Iloprost Iloprost is a stable PGI2 analogue that is
completed the transition; however, there was a modest increase delivered via an aerosolized device six to nine times per day.
in mPAP and decrease in CI. Noteworthy is that the dose of IV Iloprost was studied in a 12-week, multicenter, placebo-controlled,
[21/6/010/15:27:52] [98–117]
Table 10.4 Risk Assessment for PAH

Determinates of risk Lower risk (good prognosis) Higher risk (poor prognosis)
Clinical evidence of RV failure No Yes
Progression of symptoms Gradual Rapid
WHO classa II, III IV
6MW distanceb Longer (greater than 400 m) Shorter (less than 300 m)
CPET Peak VO2 greater than 10.4 mL/kg/min Peak VO2 less than 10.4 mL/kg/min
Echocardiography Minimal RV dysfunction Pericardial effusion, significant RV enlargement/dysfunction,
right atrial enlargement
Hemodynamics RAP less than 10 mmHg, CI greater RAP greater than 20 mmHg, CI less than 2.0 L/min/m2
than 2.5 L/min/m2
BNPc Minimally elevated Significantly elevated
Note: Most data available pertains to IPAH. Little data is available for other forms of PAH. One should not rely on any single factor to make risk
predictions.
a
WHO class is the functional classification for PAH and is a modification of the New York Heart Association functional class.
b
6MW distance is also influenced by age, gender, and height.
c
As there is currently limited data regarding the influence of BNP on prognosis, and many factors including renal function, weight, age, and
gender may influence BNP, absolute numbers are not given for this variable.
Abbreviations: 6MW, six-minute walk; BNP, brain natriuretic peptide; CI, cardiac index; CPET, cardiopulmonary exercise testing; peak VO2,
average peak oxygen uptake during exercise; RAP, right atrial pressure; RV, right ventricle; WHO, World Health Organization; PAH,
pulmonary arterial hypertension.
Source: Reprinted from Refs. 16 and 68.
randomized trial of 207 FC III and IV patients with either hormonal contraception so women of child-bearing age must
IPAH, PAH associated with CTD or appetite suppressants, be counseled to use dual contraception for birth control. Other
or PH related to inoperable chronic thromboembolic disease side effects include headache, flushing, lower-extremity edema,
(122). Treatment with iloprost resulted in meeting a novel and anemia. Treatment with bosentan requires monitoring of
composite end point of improvement in FC by at least liver function tests on a monthly basis, and pregnancy tests on
one level and increase in 6MWD by at least 10% in the absence women of child-bearing potential on a monthly basis, and
of clinical deterioration (16.8% vs. 4.9%, treated vs. placebo, hemoglobin/hematocrit on a quarterly basis. Patients should
P ¼ 0.007). It was generally well tolerated with coughing, be counseled regarding potential for lower-extremity edema,
headache and flushing occurring as most common side effects especially in the initial weeks of therapy, and possible need for
(Table 10.5). diuretic adjustments. Glyburide and cyclosporine A are contra-
indicated with bosentan because of significant drug-drug inter-
actions (Table 10.5).
Endothelin Receptor Antagonists
Bosentan Ambrisentan
Bosentan, a nonselective endothelin receptor blocker, was the Ambrisentan is a selective ETA receptor antagonist studied in
first orally available therapy approved for PAH. It was studied two placebo-controlled, randomized, 12-week study of WHO
in two placebo-controlled, randomized trials of FC III or IV Group I patients (ARIES-1 and ARIES-2), which were con-
patients with either PAH or PAH related to CTD (123,124). In ducted in the United States and Europe/South America,
the pivotal BREATHE-1 trial, bosentan improved the primary respectively (127). The treatment resulted in a significant
end point of 6MWD by 36 m, whereas placebo patients deter- improvement in 6MWD and delay in time to clinical worsening
iorated by 8 m (P ¼ 0.0002). Bosentan also improved the in all treatment groups. Ambrisentan is available in 5-mg and
composite end point of time to clinical worsening, which was 10-mg oral tablets taken once a day.
defined as death, initiation of IV epoprostenol, hospitalization The incidence of hepatic transaminase elevation >3 times
for worsening PAH, lung transplantation or atrial septostomy. the upper limit of normal was 0.8% for patients receiving
Long-term observational findings in survival of patients treated ambrisentan (127). This was further investigated in a recently
with bosentan as first line therapy have shown improved published study of 36 patients, who did not tolerate bosentan
survival compared with expected outcome based on the NIH or sitaxsantan because of hepatic transaminase increases and
registry equation (125). Bosentan was shown to be effective in were placed on ambrosentan therapy (128). Ambrisentan ther-
mildly symptomatic patients in EARLY study, a six-month, apy was tolerated well in this group. Peripheral edema is
multicenter, placebo-controlled trial, which enrolled 168 FC II another side effect of the ERA class and was reported in mild
PAH patients (126). The results demonstrated a significant to moderate severity in the clinical trials (127). An increase
decrease in PVR, which was the primary end point to evaluate report of incidences of peripheral edema during post marketing
treatment effects on vascular remodeling, and a significant use has prompted the FDA to issue a labeled warning for
delay in clinical worsening. elderly patients (129). The mechanisms behind this observed
Bosentan is mainly metabolized through the hepatic P450 edema is currently undergoing evaluation. No drug interaction
enzymes and increase in hepatic transaminases >3 times the was found with sildenafil (129). Ambrisentan is teratogenic.
upper limit of normal has been reported in 10% to 12% (124). Monthly blood test for liver function tests and pregnancy tests
Bosentan is teratogenic and may decrease the efficacy of for women of child-bearing age is required (Table 10.5).
[21/6/010/15:27:52] [98–117]
Table 10.5 Approved Treatments for PAH and Commonly Reported Side Effects
Drug, class Dose/route of administration Common side effects Comments
Epoprostenol (Flolan), Start 2 ng/kg/min intravenously and Prostanoid side effectsa Central tunneled catheter needed.
prostanoid titrate following side effect and Long-term data available.
PAH symptoms Effective in advanced PAH.
Dosing individualized Because of therapy being complicated, it is
Optimal dose for chronic therapy recommended that patients be referred to
25–40 ng/kg/min specialist PAH centers.
Treprostinil Start 1.25–2.5 ng/kg/min Prostanoid side effectsa Injection site pain affects majority of
(Remodulin), subcutaneously and titrate Injection site pain and patients.
prostanoid following side effect and PAH erythema Experienced centers have reported
symptoms successful outcome in managing patients
with injection site pain.
Long-term survival data available.
Treprostinil Start 2 ng/kg/min intravenously and Prostanoid side effectsa More convenient for chronic infusion than
(Remodulin), titrate following side effect and Leg pain (more common epoprostenol.
prostanoid PAH symptoms. than with epoporstenol) Revised recommendations for central
tunneled catheter care (see text).
Treprostinil (Tyvaso), 6 mcg per inhalation; titrate to target Prostanoid side effectsa Recent approval
prostanoid maintenance dosage of nine (generally less severe) Rapid inhalation time.
inhalations and cough
Four inhaled treatments daily
Iloprost (Ventavis), 5 mcg per inhalation Prostanoid side effectsa Well tolerated with oral PAH therapies.
prostanoid 6–9 inhaled treatments daily (generally less severe) Compliance can be an issue with need for
and cough frequent treatments.
Recent approval of 20 mcg/mL concentration
to decrease treatment duration.
Bosentan (Tracleer), 62.5 mg twice daily oral 4 wk, then Headache, dizziness, Monthly LFTs required.
ERA 125 mg twice daily if LFT normal edema Contraindicated with cyclosporine and
glyburide.
Decreases effectiveness of oral hormonal
contraceptives.
Longer-term observational survival data
available.
Decrease in pulmonary vascular resistance
in FC II patients.
Ambrisentan 5 or 10 mg once daily oral Peripheral edema, nasal Monthly LFTs required, but lower incidence
(Letairis), ERA congestion, sinusitis of LFT abnormalities compared with other
ERAs.
Higher incidence of edema in elderly
patients compared with other ERAs.
May decreases effectiveness of oral
hormonal contraceptives.
Longer-term observational survival data
available.
No drug interaction observed in combined
treatment with sildenafil.
Sildenafil (Revatio), 20 mg thrice daily oral Epistaxis, headache, Contraindicated with nitrates.
PDE5-Inh flushing, diarrhea Some patients may need up titration of
dose.
Tadalafil (Adcirca), 40 mg once daily oral Headache, myalgia, Contraindicated with nitrates.
PDE5-Inh flushing
a
Side effects related to prostacyclin: jaw pain, diarrhea, flushing, headache, nausea.
Abbreviations: ERA, endothelin receptor antagonists; LFT, liver function test; PAH, pulmonary arterial hypertension; PDE5-Inh, phospho-
diesterase-5 inhibitor.
Phosphodiesterase-5 Inhibitor however, nearly all patients were titrated up to a dose of 80 mg

Sildenafil three times a day. Side effects include headache, flushing,
Sildenafil was studied in a 12-week randomized placebo- dyspepsia and epistaxis (Table 10.5).
controlled study of 278 symptomatic PAH patients (130). The
primary end point of 6MWD improved by 45, 46, and 50 m in Tadalafil
the 20-, 40-, and 80-mg groups, respectively (P < 0.001). There Tadalafil, a PDE5-Inh with a longer half-life than sildenafil, was
was no change in the time to clinical worsening at week 12. The recently studied in a 16-week, double-blind, placebo-controlled
result of 222 patients who completed one year of treatment trial among 405 PAH patients using 2.5-, 10-, 20-, and 40-mg
demonstrated that the 6MWD improvement was maintained; tablets once a day (131). The highest dose of tadalafil
[21/6/010/15:27:52] [98–117]
demonstrated a 41 m increase in 6MWD compared with 9 m for The recommended targeted to international normalized ratio
placebo (P < 0.001). There was also a delay in the time to varies from 2.0 to 2.5 and 2.0 to 3.0 to 1.5 to 2.0 in some centers.
clinical worsening (defined as death, hospitalization, initiation In patients with APAH, anticoagulation is controversial with
of new PAH therapy, worsening WHO FC). Side effects include few data to support its use. In CTD and portopulmonary
headache, diarrhea, nausea, back pain, dizziness, dyspepsia patients, the risk of gastrointestinal bleeding may be increased.
and flushing (Table 10.5). Most experts recommend warfarin anticoagulation in APAH
patients being treated with IV prostanoids in the absence of
contraindicating factors.
Conventional Treatment Hypoxemia is a potent pulmonary vasoconstrictor and
CCBs are recommended for patients who demonstrate respon- thus can contribute to progression of PAH. It is recommended
siveness during an acute vasodilator testing (see Acute Vaso- that patients with PAH to maintain oxygen saturation >90% at
dilator Test). Patients with IPAH who meet the criteria may be all times though the use of supplemental oxygen in patients
considered treatment with CCBs. Long acting nifedipine, dil- with Eisenmenger physiology is controversial. Diuretics are
tiazem, or amlodipine is suggested. Verapamil should be used to treat volume overload because of right heart failure.
avoided because of its potential negative inotropic effects. For diuretic naı̈ve patients, slow initiation and monitoring of
Patients need to be followed closely for efficacy and safety on renal function are recommended with goal of attaining near-
CCBs. If a patient does not improve to FC I or II with CCBs, the normal intravascular volume. In acute decompensated right
patient should not be considered a chronic responder and heart failure and/or in presence of diuretic resistance, IV
PAH-directed treatment should be initiated. diuretics are needed. Although digoxin has not been well
Anticoagulation has been studied in two small uncon- studied in patients with PAH, it is used with careful monitoring
trolled trials in IPAH patients. On the basis of these studies, in low doses in the setting of refractory right heart failure and/
most experts recommend warfarin anticoagulation (68,132). or atrial arrhythmia.
Figure 10.7 PAH evidence-based treatment algorithm. Drugs within the same grade of evidence are listed in alphabetical order and not
order of preference. Not all agents listed are approved or available for use in all countries. *To maintain oxygen at 92%. +Investigational,
under regulatory review. Abbreviations: APAH, associated pulmonary arterial hypertension; ERA, endothelin receptor antagonist; HPAH,
heritable pulmonary arterial hypertension; IPAH, idiopathic pulmonary arterial hypertension; IV, intravenous; PAH, pulmonary arterial
hypertension; PDE-5, phosphodiesterase type 5; SC, subcutaneous; WHO, World Health Organization. Source: From Ref. 9.
[21/6/010/15:27:52] [98–117]
Combination Therapy in PAH placebo. At the end of 16 weeks, more than 80% of patients had
With the approval of therapies targeting different pathways, reached the 80-mg TID dosing level. The primary end point
utilizing combination approach in the hope of improving out- was change in 6MWD and there was placebo-adjusted increase
come has attracted marked interest. The potential to increase of 26 m in the subjects who received sildenafil. There were
efficacy by utilizing combination therapy must be measured seven deaths in the placebo group and none among patients
against possible toxicity and drug-drug interactions. Several receiving sildenafil. Clinical worsening events defined as death,
small, open-label observational studies reported potential ben- transplant, hospitalization, or an increase in epoprostenol dose,
efit of combination treatment (133,134). Initial study, evaluating were significantly different in favor of the treated group. Sev-
combining bosentan or placebo to FC III or IV patients receiv- eral large studies are currently underway evaluating the effect
ing IV epoprostenol, failed to show benefit thought this study of combining different classes of oral regimen including the
was underpowered (135). Two studies evaluated adding COMPASS-2 trial (Effects of combination of bosentan and
inhaled iloprost to bosentan therapy in a randomized, dou- sildenafil versus sildenafil monotherapy on morbidity and
ble-blind, placebo-controlled design. The STEP study enrolled mortality in symptomatic patients with PAH), which is the
67 patients in a 12-week study which demonstrated safety as first morbidity/mortality driven trial focusing on combination
well as improvement in 6MWD (26 m, placebo corrected, P ¼ therapy in PAH.
0.051); the COMBI study which evaluated 40 patients failed to
demonstrate benefit and the study was terminated (136,137).
The largest completed combination trial in PAH to date is Treatment Algorithm and Assessing
PACES study, which added adding sildenafil as add-on ther- Response to Therapy
apy to IV epoprostenol (138). This 16-week, multinational, The most recent treatment guideline from the Dana Point
double-blind, placebo-controlled study enrolled 267 patients meeting is shown in Figure 10.7 (9). Incorporating risk-based
who were on stable epoprostenol therapy. Patients were approach by combining known factors that determine progno-
randomized to receive 20 mg three times a day titrated to sis in PAH in selecting therapy has been recommended and is
40- and 80-mg TID, at four week intervals, or corresponding widely utilized by clinicians (Fig. 10.8) (16). The ACCF/AHA
Figure 10.8 PAH treatment algorithm based on risk assessment. Background therapies include warfarin anticoagulation, which is
recommended in all patients with IPAH without contraindication. Diuretics are used for management of right heart failure. Oxygen is
recommended to maintain oxygen saturation greater than 90%.aAcute vasodilator testing should be performed in all IPAH patients who may
be potential candidates for long-term therapy with CCBs. Patients with PAH due to conditions other than IPAH have a very low rate of long-
term responsiveness to oral CCBs, and the value of acute vasodilator testing in such patients needs to be individualized. IPAH patients in
whom CCB therapy would not be considered, such as those with right heart failure or hemodynamic instability, should not undergo acute
vasodilator testing. aCCBs are indicated only for patients who have a positive acute vasodilator response, and such patients need to be
followed closely for both safety and efficacy. bFor patients who did not have a positive acute vasodilator testing and are considered lower risk
on the basis of clinical assessment (Table 10.4), oral therapy with endothelin receptor antagonist (ERA) or phosphodiesterase-5 inhibitor
(PDE-5 I) would be the first line of therapy recommended. If an oral regimen is not appropriate, the other treatments would need to be
considered on the basis of patient’s profile and the side effects and risk of each therapy. cFor patients who are considered high risk on the
basis of clinical assessment (Table 10.4), continuous treatment with intravenous (IV) prostacyclin (epoprostenol or treprostinil) would be the
first line of therapy recommended. Combination therapy should be considered when patients are not responding adequately to initial
monotherapy. Timing for lung transplantation and/or atrial septostomy is challenging and is reserved for patients who progress despite
optimal medical treatment. Abbreviations: IPAH, idiopathic pulmonary arterial hypertension; PAH, pulmonary arterial hypertension; CCBs,
calcium channel blockers. Source: From Ref. 68.
[21/6/010/15:27:52] [98–117]
Table 10.6 Longitudinal Evaluation of Pulmonary Arterial Hypertension Patients on Therapy

Clinical course Stable Unstable
. No increase in symptoms and/or decompensation . Increase in symptoms and/or decompensation
. No evidence of right heart failure . Signs of right heart failure
. FC I/II . FC IVa
. 6MWD > 400 m . 6MWD < 300 ma
. RV size/function normal . RV enlargement/dysfunction
. RAP normal; CI normal . RAP high; CI low
. BNP normal/stable or decreasing . BNP elevated/increasing
. Oral therapy . IV prostacyclin and/or combination treatment
Frequency of evaluation Q 3–6 mob Q 1–3 mo
FC assessment Every clinic visit Every clinic visit
6MWD Every clinic visit Every clinic visit
Echocardiogramc Q 12 mo or center dependent Q 6–12 mo or center dependent
BNPd Center dependent Center dependent
Right heart catheterization Clinical deterioration and center dependent Q 6–12 mo or clinical deterioration
Note: For patient in the high-risk category, consider referral to a PH specially center for consideration for advanced therapies, clinical trials,
and/or lung transplantation.
a
The frequency of follow-up evaluation for patient in FC III and/or 6MWD between 300 and 400 m would depend on composite of detailed
assessments on other clinical and objective characteristics listed.
b
For patient who remain stable on established therapy, follow-up assessments can be performed by referring physican(s) or PH specialty
centers.
c
Echocardiographic measurement of PASP is estimation only and it is strogly advised not to rely on its evaluation as the sole parameter to
make therapeutics decisions.
d
The utility of serial BNP levels to guide management in individual patients has not established.
Abbreviations: BNP, brain natriuretic peptide; CI, cardiac index; FC, functional class; 6MWD, six-minute walk distance; Q, every; RAP, right
atrial pressure; PH, pulmonary hypertension; RV, right ventricle; IV, intravenous; mo, months.
2009 Expert Consensus Document recently published a guide- 4. Hatano S, Strasser R, eds. Primary Pulmonary Hypertension.
line outlining recommendations in assessing response and fol- Geneva: World Health Organization, 1975.
lowing patients on therapy (Table 10.6) (68). 5. Rich S, Dantzker DR, Ayres SM, et al. Primary pulmonary hyper-
tension: a national prospective study. Ann Intern Med 1987; 107:
216–223.
6. D’Alonzo GE, Barst RJ, Ayres SM, et al. Survival in patients with
CONCLUSION primary pulmonary hypertension: results from a national pro-
The field of PAH has advanced from a disease of little knowl- spective registry. Ann Intern Med 1991; 115:343–349.
edge and treatment to improved understanding of the pathobi- 7. Simonneau G, Nazzareno G, Rubin LJ, et al. Clinical classification
ology with seven approved drugs targeting known identified of pulmonary hypertension. J Am Coll Cardiol 2004; 43 (suppl 1,
pathways. As recognition and screening of PH increases, accu- issue 12):S5–S12.
rate diagnosis of PAH utilizing the proper techniques of RHC 8. Simonneau G, Robbins IM, Beghetti M, et al. Updated clinical
classification of pulmonary hypertension. J Am Coll Cardiol
takes greater significance. Continued efforts are being pursued
2009; 54:S43–S54.
in several directions: to study other effected pathways in PAH 9. Barst RJ, Gibbs SR, Hossein A, et al. Updated evidence-based
and novel therapies to targeting the specific pathways; to treatment algorithm in pulmonary arterial hypertension. J Am
update current epidemiology and treatment patterns; ongoing Coll Cardiol 2009; 54:S78–S84.
studies evaluating the efficacy and safety of using combination 10. Grossman W, Barry WH. Cardiac catheterization. In: Braunwald
therapies in this complex disease. The study of patients being E, ed. Heart Disease: a Textbook of Cardiovascular Medicine.
affected with disproportionate PH in association with underly- Philadelphia: WB Saunders, 1988, 247–252.
ing hypoxic lung disease, left-sided heart disease and chronic 11. Badesch DB, Champion HC, Sanchez MA, et al. Diagnosis and
thromboembolic disease need to be further elucidated. assessment of pulmonary arterial hypertension. J Am Coll Car-
diol 2009; 54:S55–S66.
12. Yuan JXJ, Rubin LJ. Pathogenesis of pulmonary arterial hyper-
tension: the need for multiple hits. Circulation 2005; 111:534–538.
REFERENCES 13. Machado RD, James V, Southwood M, et al. Investigation of a
1. Romberg E. Ueber sklerose der lungen arterie. Dsch Arch Klin second genetic hits at the BMPR2 locus as a modulator of disease
Med 1891; 48:197–206. progression in familial pulmonary arterial hypertension. Circu-
2. Dresdale DT, Schultz M, Michtom RJ. Primary pulmonary hyper- lation 2005; 111:607–613.
tension: I. Clinical and hemodynamic study. Am J Med 1951; 11: 14. Christman BW, McPherson CD, Newman JH, et al. An imbalance
686–705. between the excretion of thromboxane and prostacyclin metab-
3. Wood P, Besterman EM, Towers MK, et al. The effect of acetyl- olites in pulmonary hypertension. N Engl J Med 1992; 327:70–75.
choline on pulmonary vascular resistance and left atrial pressure 15. Tuder RM, Cool CD, Geraci MW, et al. Prostacyclin synthase
in mitral stenosis. Br Heart J 1957; 19:276–286. expression is decreased in lungs from patients with severe
[21/6/010/15:27:52] [98–117]
pulmonary hypertension. Am J Respir Crit Care Med 1999; Institute Working Group on Cellular and Molecular Mechanisms
159:1925–1932. of Right Heart Failure. Circulation 2006; 114:1883–1891.
16. McLaughlin VV, McGoon MD. Pulmonary arterial hypertension. 36. Dias CA, Assad RS, Caneo LF, et al. Reversible pulmonary
Circulation 2006; 114:1417–1531. trunk banding, II: an experimental model for rapid pulmonary
17. Yanagisawa M, Kurihara H, Kimura S, et al. A novel potent ventricular hypertrophy. J Thorac Cardiovasc Surg 2002; 124:
vasoconstrictor peptide produced by vascualar endothelial cells. 999–1006.
Nature 1988; 332:411–415. 37. Louie EK, Lin SS, Reynertson SI, et al. Pressure and volume
18. Ogawa Y, Nakao K, Arai H, et al. Molecular cloning of a non- loading of the right ventricle have opposite effects on left
isopeptide-selective human endothelin receptor. Biochem Bio- ventricular ejection fraction. Circulation 1995; 92:819–824.
phys Res Commun 1991; 178:248–255. 38. Hopkins WE, Waggoner AD. Severe pulmonary hypertension
19. Giad A, Yanagisawa M, Langleben D, et al. Expression of without right ventricular failure: the unique hearts of patients
endothelin-1 in the lungs of patients with pulmonary hyperten- with Eisenmenger syndrome. Am J Cardiol 2002; 89:34–38.
sion. N Engl J Med 1993; 328:1732–1739. 39. Busjahn A, Knoblauch H, Knoblauch M, et al. Angiotensin-
20. Rubens C, Ewert R, Halank M, et al. Big endothelin-1 and converting enzyme DD genotype in patients with primary pul-
endothelin-1 plasma levels are correlated with the severity of monary hypertension: increased frequency and association with
primary pulmonary hypertension. Chest 2001; 120:1562–1569. preserved hemodynamics. J Renin Angiotensin Aldosterone Syst
21. Giaid A, Saleh D. Reduced expression of endothelial nitric oxide 2003; 4:27–30.
synthase in the lungs of paients with pulmonary hypertension. N 40. Masuyama T, Kodama K, Kitakabe A, et al. Continuous-wave
Engl J Med 1995; 333:214–221. Doppler echocardiographic detection of pulmonary regurgitation
22. Lee SL, Wang WW, Lanzillo JJ, et al. Serotonin produces both and its application to noninvasive estimation of pulmonary
hyperplasia and hypertrophy of bovine pulmonary artery artery pressure. Circulation 1986; 74:484–492.
smooth muscle cell in culture. Am J Physiol 1994; 266:L46–L52. 41. Weir EK, Michelakis ED, Archer SL, et al. Pulmonary hyperten-
23. Herve P, Launay JM, Scrobohaci ML, et al. Increased plasma sion. In: Willerson JT, Cohn JN, eds. Cardiovascular Medicine.
serotonin in primary pulmonary hypertension. Am J Med 1995; 2nd ed. Philadelphia: Churchill Livingstone, 2000:1856–1884.
99:249–254. 42. Kircher BJ, Himelman RB, Schiller NB. Noninvasive estimation of
24. Abenhaim L, Moride Y, Brenot F, et al. Appetite-suppressant right atrial pressure from the inspiratory collapse of the inferior
drugs and the risk of primary pulmonary hypertension. N Engl J vena cava. Am J Cardiol 1990; 15:493–496.
Med 1996; 335:609–616. 43. Hinderliter AL, Willis PW IV, Long WA, et al. Frequency and
25. Eddahibi S, Humbert M, Fadel E, et al. Serotonin transporter over severity of tricuspid regurgitation determined by Doppler echo-
expression is responsible for pulmonary artery smooth muscle cardiography in primary pulmonary hypertension. Am J Cardi-
hyperplasia in primary pulmonary hypertension. J Clin Invest olol 2003; 91:1033–1037.
2001; 108:1141–1150. 44. Currie PJ, Seward JB, Chan KL, et al. Continuous wave Doppler
26. Marcos E, Adnot S, Pham MH, et al. Serotonin transporter determination of right ventricular pressure: a simultaneous
inhibitors protect against hypoxic pulmonary hypertension. Am Doppler-catheterization study in 127 patients. J Am Coll Cardiol
J Respir Crit Care Med 2003; 168:487–493. 1985; 6:750–756.
27. Yuan JJ, Aldinger AM, Juhaszova M, et al. Dysfunctional voltage- 45. Hinderliter AL, Willis PW, Barst RJ, et al. Effects of long-term
gated Kþ channels in pulmonary artery smooth muscle cells of infusion of prostacyclin (epoprostenol) on echocardiographic
patients with primary pulmonary hypertension. Circulation measures of right ventricular structure and function in primary
1998; 98:1400–1406. pulmonary hypertension: Primary Pulmonary Hypertension
28. Weir EK, Reeve HL, Huang JM, et al. Anorexic agents aminorex, Study Group. Circulation 1997; 95:1479–1486.
fenfluramine, and dexfenfluramine inhibit potassium current in 46. McQuillan BM, Picard MH, Leavitt M, et al. Clinical correlates
rat pulmonary vascular smooth muscle and cause pulmonary and reference intervals for pulmonary artery systolic pressure
vasoconstriction. Circulation 1996; 94:2216–2230. among echocardiographically normal subjects. Circulation 2001;
29. Welsh CH, Hassell KL, Badesch DB, et al. Coagulation and 104:2797–2802.
fibrinolytic profiles in patients with severe pulmonary hyperten- 47. Bossone E, Rubenfire M, Bach DS, et al. Range of tricuspid
sion. Chest 1996; 110:710–717. regurgitation velocity at rest and during exercise in normal
30. Tuder RM, Groves BM, Badesch DB, et al. Exuberant endothelial adult men: implications for the diagnosis of pulmonary hyper-
cell growth and elements of inflammation are present in plexi- tension. J Am Coll Cardiol 1999; 33:1662–1666.
form lesions in pulmonary hypertension. Am J Pathol 1994; 48. Chan KL, Currie PJ, Seward JB, et al. Comparison of three
144:275–285. Doppler ultrasound methods in the prediction of pulmonary
31. Lane KB, Machado RD, Pauciulo MW, et al. Heterozygous germ- artery pressure. J Am Coll Cardiol 1987; 9:549–554.
line mutations in BMPR2, encoding a TGF-b receptor, cause 49. Yock PG, Popp RL. Noninvasive estimation of right ventricular
familial primary pulmonary hypertension: the International systolic pressure by Doppler ultrasound in patients with tricus-
PPH Consortium. Nat Genet 2000; 26:81–84. pid regurgitation. Circulation 1984; 70:657–662.
32. Deng Z, Morse JH, Slager SL, et al. Familial primary pulmonary 50. Brecher SJ, Gibbs JS, Fox KM, et al. Comparison of Doppler
hypertension (gene PPH1) is caused by mutations in the bone derived hemodynamic variables and stimultaneous high fidelity
morphogenetic protein receptor-II gene. Am J Hum Genet 2000; pressure measurements in severe pulmonary hypertension. Br
67:737–744. Heart J 1994; 72:384–389.
33. Trembath R, Thomson J, Machado R, et al. Clinical and molecular 51. Arcasoy SM, Christie JD, Ferrari VA, et al. Echocardiographic
genetic features of pulmonary hypertension in patients with assessment of pulmonary hypertension in patients with
hereditary hemorrhagic telangiectasia. N Engl J Med 2001; advanced lung disease. Am J Respir Crit Care Med 2003; 167:
345:325–334. 735–740.
34. Kukulski T, Hubbert L, Arnold M, et al. Normal regional right 52. Homma A, Anzueto A, Peters JI, et al. Pulmonary artery systolic
ventricular function and its changes with age: a Doppler pressures estimated by echocardiogram vs cardiac catheterization
myocardial imaging study. J Am Soc Echocardiogr 2000; 13: in patients awaiting lung transplantation. J Heart Lung Transplant
194–204. 2001; 20:833–839.
35. Voekel NF, Quaife RA, Leinwant LA, et al. Right ventricular 53. Tramarin R, Torbicki A, Marchandise B, et al. Doppler echocar-
function and failure. Report of a National Heart, Lung, and Blood diographic evaluation of pulmonary artery pressure in chronic
[21/6/010/15:27:52] [98–117]
obstructive pulmonary disease. A European multicenter study. 73. Kitzman DW, Little WC, Brubaker PH, et al. Pathophysiological
Eur Heart J 1991; 12:103–111. characterization of isolated diastolic heart failure in comparison
54. Denton CP, Cailes JB, Phillips GD, et al. Comparison of Doppler to systolic heart failure. JAMA 2002; 288:2144–2150.
echocardiography and right heart catheterization to assess pul- 74. Zile MR, Brutsaert DL. New concepts in diastolic dysfunction
monary hypertension in systemic sclerosis. Br J Rheumatol 1997; and diastolic heart failure. Part II: causal mechanisms and treat-
36:239–243. ment. Circulation 2002; 105:1503–1508.
55. Pepine CJ, Hill JA, Lambert CR. Pressure measurement and 75. Ghio S, Gavazzi A, Campana C, et al. Independent and additive
determination of vascular resistance. In: Pepine CJ, Hill JA, prognostic value of right ventricular systolic function and pul-
Lambert CR, eds. Diagnostic and Therapeutic Cardiac Catheter- monary artery pressure in patients with chronic heart failure. J
ization. 2nd ed. Philadelphia: Williams & Wilkins, 1994:355–371. Am Coll Cardiol 2001; 37:183–188.
56. Ragosta M. Right sided heart disorders. In: Ragosta M, ed. 76. Hoeper MM, Barbera JA, Channick RN, et al. Diagnosis, assess-
Textbook of Clinical Hemodyanmics. Philadelphia: Saunders, ment, and treatment of non-pulmonary arterial hypertension
Inc., 2008:109–122. pulmonary hypertension. J Am Coll Cardiol 2009; 54:S85–S96.
57. Ragosta M. Normal waveforms, artifacts, and pitfalls. In: Ragosta 77. Califf RM, Adams KF, McKenna WJ, et al. A randomized con-
M, ed. Textbook of Clinical Hemodynamics. Philadelphia: Saun- trolled trial of epoprostenol therapy for severe congestive heart
ders, Inc., 2008:16–37. failure: the Flolan International Randomized Survival Trial
58. Pichard AD, Kay R, Smith H, et al. Large V waves in the (FIRST). Am Heart J 1997; 134:44–54.
pulmonary wedge pressure tracing in the absence of mitral 78. Mylona P, Cleland JG. Update of REACH-1 and MERIT-HF
regurgitation. Am J Card 1982; 50:1044–1050. clinical trials in heart failure. Eur J Heart Fail 1999; 1:197–200.
59. Fuchs RM, Henser RR, Yin FCP, et al. Limitation of pulmonary 79. Teerlink JR. Recent heart failure trials of neurohormonal modu-
wedge V waves in diagnosing mitral regurgitation. Am J Cardiol lation (OVERTURE and ENABLE): approaching the asymptote of
1982; 49:849–854. efficacy? J Card Fail 2002; 8:124–127.
60. Sandoval J, Baurle O, Palomar A, et al. Survival in primary 80. Lewis GD, Semigran MJ. The emerging role for type 5 phospho-
pulmonary hypertension: validation of a prognostic equation. diesterase inhibition in heart failure. Curr Hear Fail Rep 2006;
Circulation 1999; 89:1733–1744. 3:123–128.
61. McLaughlin VV, Presberg KW, Doyle RL, et al. American College 81. Salloum FN, Abbate A, Das A, et al. Sildenafil (Viagra) attend-
of Chest Physicians. Prognosis of Pulmonary arterial hyperten- uates ischemic cardiomyopathy and improves left ventricular
sion: ACCP evidence-based clinical practice guidelines. Chest function in mice. Am J Physiol Heart Circ Phyiol 2008; 294:
2004; 126:78S–92S. H1398–H1406.
62. Sitbon O, Humbert M, Nunes H, et al. Long-term intravenous 82. Lewis GD, Lachmann J, Camuso J, et al. Sildenafil improves
epoprostenol infusion in primary pulmonary hypertension: prog- exercise hemodynamics and oxygen uptake in patients with
nostic factors and survival. J Am Coll Cardiol 2002; 40:780–788. systolic heart failure. Circulation 2007; 115:59–66.
63. Groves BM, Badesch DB, Turkevich D, et al. Correlation of acute 83. Lewis GD, Shah R, Shahzad K, et al. Sildenafil improves exercise
prostacyclin response in primary (unexplained) pulmonary capacity and quality of life in patients with systolic heart failure
hypertension with efficacy of treatemt with calcium channel and secondary pulmonary hypertension. Circulation 2007;
blockers and survival. In: Weir K, ed. Ion Flux in Pulmonary 116:1555–1562.
Vascular Control. New York: Plenum Press, 1993:317–330. 84. Butler J, Chomsky DB, Wilson JR. Pulmonary hypertension and
64. Sitbon O, Brenot F, Denjean A, et al. Inhaled nitric oxide as a exercise intolerance in patient with heart failure. J Am Coll
screening vasodilator agent in primary pulmonary hypertension: Cardiol 1999; 34:1802–1806.
a dose-response study and comparison with prostacyclin. Am J 85. Kingsbury MP, Huang W, Donnelly JL, et al. Structural remodel-
Respir Crit Care med 1995; 151:384–389. ing of lungs in chronic heart failure. Basic Res Cardiol 2003;
65. Schrader BJ, Inbar S, Kaufmann L, et al. Comparison of the effects 98:295–303.
of adenosine and nifedipine in pulmonary hypertension. J Am 86. Todorovich-Hunter L, Dodo H, Ye C, et al. Increased pulmnary
Coll Cardiol 1992; 19:1060–1064. artery elastolytic activity in adult rats with monocrotaline
66. Sitbon O, Humbert M, Jagot JL, et al. Inhaled nitric oxide as a induced progressive hypertensive pulmonary vascular disease
screening agent for safely identifying responders to oral calcium- compared with infant rats with nonprogressive disease. Am Rev
channel blockers in primary pulmonary hypertension. Eur Respir Respir Dis 1992; 146:213–223.
J 1998; 12:265–270. 87. Chen JM, Levin HR, Michler RE, et al. Reevaluating the signif-
67. Sitbon O, Humbert M, Jais X, et al. Long-term response to icance of pulmonary hypertension before cardiac transplantation:
calcium channel blockers in idiopathic pulmonary arterial hyper- determination of optimal thresholds and quantification of the
tension. Circulation 2005; 111:3105–3111. effect of reversibility on perioperative mortality. J Thorac Car-
68. McLaughlin VV, Archer SL, Badesch DB, et al. ACCF/AHA 2009 diovasc Surg 1997; 114:627–634.
expert consensus document on pulmonary hypertension: a 88. Taylor DO, Edwards LB, Aurora P, et al. Registry of the Inter-
report of the American College of Cardiology Foundation Task national Society for Heart and Lung Transplantation: twenty-
Force on Expert Consensus Documents. J Am Coll Cardiol 2009; fifth official adult heart transplantation report—2008. J Heart
53:1573–1619. Lung Transplant 2008; 27:943–956.
69. Palmer SM, Robinson LJ, Wang A, et al. Massive pulmonary 89. Kormos RL. Griffith BP, Hardestry RL, et al. Utility of preoper-
edema and death after prostacyclin infusion in a patient with ative right heart catheterization data as a predictor of survival
pulmonary veno-occlusive disease. Chest 1998; 113:237–240. after heart transplantation. J Heart Lung Transplant 1986; 5:391.
70. Hoeper MM, Lee SH, Voswinckel R, et al. Complications of right 90. Tsai FC, Marelli D, Bresson J, et al. Recent trends in early
heart catheterization procedures in patients with pulmonary outcomes of adult patients after transplantation: a single-institu-
hypertension in experienced centers. J Am Coll Cardiol 2006; tion review of 251 transplants using standard donor organs. Am J
48:2546–2552. Transplant 2002; 2:539–545.
71. Senni M, Tribouilloy CM, Rodeheffer RJ, et al. Congestive heart 91. Radovancevic B, Vrtovec B, Thomas CD, et al. Nitric oxide versus
failure in the community: a study of all incident cases in Olmsted prostaglandin E1 for reduction of pulmonary hypertension in
County, Minnesota, in 1991. Circulation 1998; 98:2282–2289. heart transplant candidates. J Heart Lung Transplant 2005;
72. Higg K, Swedberg K, McMurray J. Heart failure with preserved 24:690–695.
left ventricular systolic function. J Am Coll Cardiol 2004; 43: 92. Natale ME, Pina IL. Evaluation of pulmonary hypertension in
317–327. heart transplant candidates. Curr Opin Cardiol 2003; 18:136–140.
[21/6/010/15:27:52] [98–117]
93. Zimpler D, Zrunek P, Roethy W, et al. Left ventricular assist 114. Aguilar RV, Farber HW. Epoprostenol (prostacyclin) therapy in
devices decrease fixed pulmonary hypertension in cardiac trans- HIV-associated pulmonary hypertension. Am J Respir Crit Care
plant candidates. J Thorac Cardiovasc Surg 2007; 133:689–695. Med 2000; 162:1846–1850.
94. Klotz S, Deng MC, Stypmann J, et al. Left ventricular pressure and 115. Rosenzweig EB, Kerstein D, Barst RJ. Long-term prostacyclin for
volume unloading during pulsatile versus nonpulsatile left ven- pulmonary hypertension with associated congenital heart
tricular assist device support. Ann Thorac Surg 2004; 77:143–149. defects. Circulation 1999; 99:1858–1865.
95. Garatti A, Bruschi G, Colombo T, et al. Is fixed severe pulmonary 116. Kuo PC, Johnson LB, Plotkin JS, et al. Continuous intravenous
hypertension still a contraindication to heart transplant in the infusion of epoprostenol for the treatment of portopulmonary
modern era of mechanical circulatory support? A review. J hypertension. Transplantation 1997; 63:604–616.
Cardiovasc Med 2008; 9:1059–1062. 117. Rich S, McLauglin VV. The effects of chronic prostacyclin ther-
96. Cogan JD, Vnencak-Jones CL, Phillips JA III, et al. Gross BMPR2 apy on cardiac output and symptoms in primary pulmonary
gene rearrangements constitute a new cause for primary pulmo- hypertension. J Am Coll Cardiol 1999; 34:1184–1187.
nary hypertension. Genet Med 2005; 7:169–174. 118. Park MH. Pharmacotherapeutic options for pulmonary arterial
97. Thomson JR, Machado RD, Pauciulo MW, et al. Sporadic primary hypertension. Current Medical Literature: Pulmonary Hyperten-
pulmonary hypertension is associated with germline mutations sion 2009; 1:1–21.
of the gene encoding BMPR-II, a receptor member of the TGF-B 119. Simonneau G, Barst RJ, Galie N, et al. Continuous subcutaneous
family. J Med Genet 2000; 37:741–745. infusion of treprostinil, a prostacyclin analogue, in patients with
98. Koh ET, Lee P, Gladman DD, et al. Pulmonary hypertension in pulmonary arterial hypertension. Am J Respir Crit Care Med
systemic sclerosis: an analysis of 17 patients. Br J Rheum 1996; 2002; 165:800–804.
35:989–993. 120. Tapson VF, Gomberg-Maitland M, McLaughlin V, et al. Safety
99. Kuhn KP, Byrne DW, Arbogast PG, et al. Outcome in 91 consec- and efficacy of IV treprostinil for pulmonary arterial hyperten-
utive patients with pulmonary arterial hypertension receiving sion: a prospective, multicenter, open-label, 12-week trial. Chest
epoprostenol. Am J Respir Crit Care Med 2003; 167:580–586. 2006; 129:683–688.
100. Kuo PC, Plotkin JS, Johnson LB, et al. Distinctive clinical features 121. Gomberg-Maitland M, Tapson VF, Benza RL, et al. Transition
of portopulmonary hypertension. Chest 1997; 112:980–986. from intravenous epoprotenol to intravenous treprostinil in pul-
101. Colle IO, Moreau R, Godinho E, et al. Diagnosis of portopulmo- monary hypertension. Am J Respir Crit Care Med 2005; 172:
nary hypertension in candidates for liver transplantation: a pro- 1586–1589.
spective study. Hepatology 2003; 37:401–409. 122. Olschewski H, Simonneay G, Galie N, et al. Inhaled iloprost for
102. Krowka M, Plevak DJ, Findlay JY, et al. Pulmonary hyemody- severe pulmonary hypertension. N Engl J Med 2002; 347: 322–329.
namics and perioperative cardiopulmonary mortality in patients 123. Channick RN, Simonneau G, Sitbon O, et al. Effects of the dual
with portopulmonary hypertension undergoing liver transplan- endothelin-receptor antagonist bosentan in patients with pulmo-
tation. Liver Transpl 2000; 6:443–450. nary hypertension: a randomized placebo-controlled study. Lan-
103. Walker AM, Langleben D, Korelitz JJ, et al. Temporal trends and cet 2001; 358:1119–1123.
drug exposures in pulmonary hypertension: an American expe- 124. Rubin LJ, Badesch DB, Barst RJ, et al. Bosentan therapy for
rience. Am Heart J 2006; 152:521–526. pulmonary arterial hypertension. N Engl J Med 2002; 346:
104. Opravil M, Pechere M, Speich R, et al. HIV-associated primary 896–903.
pulmonary hypertension: a case control study: Swiss HIV Cohort 125. McLaughlin VV, Sitbon O, Badesch DB, et al. Survival with first-
Study. Am J Respir Crit Care Med 1997; 155:990–995. line bosentan in patients with primary pulmonary hypertension.
105. McLaughlin VV, Shillington A, Rich S. Survival in primary Eur Respir J 2005; 25:244–249.
pulmonary hypertension: the impact of epoprostenol therapy. 126. Galie N, Rubin LJ, Hoeper MM, et al. Treatment of patients with
Circulation 2002; 106:1477–1482. mildly symptomatic pulmonary arterial hypertension with
106. Barst RJ, Rubin LJ, Long WA, et al. A comparison of continuous bosentan (EARLY study): a duble-blind, randomized controlled
intravenous epoprostenol (prostacyclin) with conventional therapy trial. Lancet 2008; 317:2093–2100.
for primary pulmonary hypertension. The Primary Pulmonary 127. Galie N, Olschewski H, Oudiz R, et al. Ambrosentan for the
Hypertension Study Group. N Engl J Med 1996; 334:296–302. treatment of pulmonary arterial hypertension. Results of the
107. Wensel R, Opitz CF, Anker SD, et al. Assessment of survival in ambrosentan in pulmonary arterial hypertension, randomized,
patients with primary pulmonary hypertension: importance of double-blind, placebo-controlled, multicenter, efficacy (ARIES)
cardiopulmonary exercise testing. Circulation 2002; 106:319–324. study 1 and 2. Circulation 2008; 117:3010–3019.
108. Raymond RL, Hinderliter AL, Willis PW, et al. Echocardio- 128. McGoon M, Frost AE, Oudiz RJ, et al. Ambrosentan therapy in
graphic predictors of adverse outcomes in primary pulmonary patients with pulmonary arterial hypertension who discontinued
hypertension. J Am Coll Cardiol 2002; 39:1214–1219. bosentan or sitaxsentan due to liver function test abnormalities.
109. Hinderliter A, Willis P 4th, Long W, et al. Frequency and Chest 2008; published online.
prognostic significance of pericardial effusion in primary pul- 129. Highlights of prescribing information Letairis. Available at:
monary hypertension. PPH Study Group. Am J Cardiol 1999; http://www.gilead.com/pdf/letairis_pi.pdf.
84:481–484. 130. Galie N, Ghofrani HA, Torbicki A, et al. Sildenafil Use in
110. Nagaya N, Nishikimi T, Uematsu M, et al. Plasma brain natriu- Pulmonary Arterial Hypertension (SUPER) Study Group. Silde-
retic peptide as a prognostic indicator in patients with primary nafil citrate therapy for pulmonary arterial hypertension. N Engl
pulmonary hypertension. Circulation 2000; 102:865–870. J Med 2005; 353:2148–2157.
111. Park MH, Scott RL, Uber PA, et al. Usefulness of B-type natriu- 131. Galie N, Brundage BH, Ghofrani HA, et al. Tadalafil therapy for
retic peptide as a predictor of treatment outcome in pulmonary pulmonary arterial hypertension. Circulation 2009; 119:2894–2903.
arterial hypertension. Congest Heart Fail 2004; 10:221–225. 132. Fuster V, Frye RL, Gersh BJ, et al. Primary pulmonary hyperten-
112. Andreassen A, Wergeland R, Simonson S, et al. N-terminal pro- sion: natural history and the importance of thrombosis. Circula-
B-type natriuretic peptide as an indicator of disease severity in a tion. 1984; 70:580–587.
heterogeneous group of patients with chronic precapillary pul- 133. Hoeper MM, Faulenbach C, Golpon H, et al. Combination ther-
monary hypertension. Am J Cardiol 2006; 98:525–529. apy with bosentan and sildenafil in idiopathic pulmonary arterial
113. Badesch DB, Tapson VF, McGoon MD, et al. Continuous intra- hypertension. Eur Respir J 2004; 24:1007–1010.
venous epoprostenol for pulmonary hypertension due to the 134. Ghofrani HA, Rose F, Schermuly RT, et al. Oral sildenafil as long-
scleroderma spectrum of disease: a randomized, controlled term adjunct therapy in severe pulmonary arterial hypertension.
trial. Ann Intern Med 2000; 132:425–434. J Am Coll Cardiol 2003; 42:158–164.
[21/6/010/15:27:52] [98–117]
135. Humbert M, Barst RJ, Robbins IM, et al. Combination of bosentan 137. Hoeper MM, Leuchte H, Halank M, et al. Combining inhaled
with epoprostenol in pulmonary arterial hypertension: BREATHE-2. iloprost with bosentan in patients with idiopathic pulmonary
Eur Respir J 2004; 24:353–359. arterial hypertension. Eur Respir J 2006; 28:691–694.
136. McLaughlin VV, Oudiz RJ, Frost AE, et al. Randomized study 138. Simonneau G, Rubin LJ, Nazzareno G, et al. Addition of silde-
adding inhaled iloprost to existing bosentan in pulmonary nafil to long-term intravenous epoprostenol therapy in patients
arterial hypertension. Am J Respir Crit Care Med 2006; 174: with pulmonary arterial hypertension. Ann Intern Med 2008;
1257–1263. 149:521–530.
[2/7/010/8:5:37] [118–123]
11
Valvular heart disease—measurement of valve orifice area

and quantification of regurgitation
Blase A. Carabello
INTRODUCTION valve area: Flow (F) ¼ valve area (A) flow velocity (V). Thus
In 1929, Werner Forssmann published a radiograph of the A ¼ F/V. In the echo laboratory, V is measured directly by
successful right heart catheterization of himself (1). Subse- Doppler ultrasonography. When using invasive hemodynamics,
quently Dickinson Richards and Andre Cournand used the the pressure gradient across the valve is measured and con-
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
heart catheter to make a series of seminal hemodynamic meas- verted to velocity where V ¼ 2 P1 P2 . Here, P1 P2 is the
urements in man resulting in the three men winning the Nobel mean pressure gradient and g is the acceleration due to gravity.
Prize in Medicine in 1956 for their accomplishments. In 1951, the This last term is incorporated because pressure is expressed in
Gorlin Formula (2) for calculating cardiac valve area was mmHg, which depends on the specific weight of mercury—
published, and for the next three decades invasive hemody- mass acceleration. Thus, in the catheterization laboratory
namics held sway as the gold standard for assessing cardiac accurate calculation of valve area relies on an accurately deter-
physiology and pathophysiology. mined pressure gradient, an accurate determination of flow
In the 1980s, the use of Doppler interrogation of the heart [cardiac output (CO)], and an accurate formula that relates the
made assessment of cardiac physiology easily applied non- two together.
invasively. While at first there was serious debate about
whether noninvasive measurements could be reliably substi-
tuted for invasive hemodynamics, the echocardiogram was AORTIC STENOSIS
eventually accepted then largely supplanted invasive hemody- Accurate Pressure Gradient Assessment
namics. In fact, the 2006 AHA/ACC Guidelines for the Man- To record an accurate transaortic valvular pressure gradient,
agement of Patients with Valvular Heart Disease (3) gives two properly placed catheters must be connected to two
invasive assessment of valve disease a class III recommendation properly calibrated transducers, reporting to an accurate
(not indicated and possibly harmful) when there is no doubt recording device. Alternatively, a pressure gradient may be
about the diagnosis following clinical and noninvasive evalua- recorded in patients in sinus rhythm during pullback of the
tion. On the other hand, it is a class I recommendation (indicated catheter from the left ventricle (LV) to the aorta recognizing
and beneficial) to obtain invasive hemodynamic data when the that a pullback determination is a “one-time” opportunity. If
diagnosis is unclear. Thus, we currently rely on a progressively premature beats or any technical problems arise during the
less practiced modality to help us solve our most important pullback the data are lost, necessitating recrossing the valve or
diagnostic dilemmas. The problem is compounded by a relative abandoning the procedure.
lack of teaching of hemodynamic principles in our cardiovas-
cular medicine fellowships and by computers that allow calcu- Proper Catheter Placement
lations to be made without the operator understanding the Figure 11.1 depicts the potential positions that might be used in
pitfalls of such calculations. Thus, the gold standard of cardiac recording a transvalvular pressure gradient, and most of them
diagnosis is threatened with a good deal of tarnish. In this light, create erroneous data (4). Of the two potential positions for the
the following is a summary of the modern use of invasive LV catheter, the proper one is with the lumen placed in the
hemodynamics in the assessment of valvular heart disease. body of the LV. As shown in Figure 11.2, in most patients with
aortic stenosis (AS), a subvalvular gradient usually exists
between the body of the LV and the LV outflow tract (5).
ASSESSMENT OF STENOTIC VALVES This gradient does not represent subvalvular AS but rather
Normal cardiac valves permit unidirectional circulatory flow at occurs as blood normally accelerates into the outflow tract
low resistance with equal pressure on both sides of an open which is narrower than the body of the LV. If the LV catheter
valve. Even when these valves are narrowed to 50% of their is placed in the outflow tract, or if the natural ejection of blood
normal aperture, only a small pressure gradient exists during pushes the catheter there, the true LV-aortic (Ao) gradient can
flow. However, further narrowing results in progressively be underestimated by as much as 20 to 40 mmHg. The distal
greater obstruction to flow and persistently higher and higher catheter should be placed in the proximal ascending aorta just
transvalvular pressure gradients. Because pressure gradient distal to the valve. For convenience, some operators have used
varies with flow, a gradient by itself may be an unreliable the side port on the femoral sheath to record the distal pressure.
indicator of stenosis severity. Knowledge of this concept led to This practice is fraught with difficulty and should be avoided
the use of valve area to help quantify stenosis severity. (6). By the time the pulse reaches the femoral artery (FA) the
Both invasive and noninvasive assessments of stenosis turbulent flow present above the valve has become relaminar-
severity use the same hemodynamic principle to calculate ized causing pressure recovery and a reduced pressure
[2/7/010/8:5:37] [118–123]
VALVULAR HEART DISEASE 119
Figure 11.2 Pressure tracings taken from high-fidelity microman-

ometer-tipped catheters. Both transducers are placed in the body of
the LV and no gradient exists (A). Both transducers are inside the
LV, and both record an LV pressure; however, one transducer is in
the body of the LV, while the other is in the LV outflow tract,
recording the pressure gradient between the two positions (B).
The true gradient between LV body and Ao is recorded (C).
Abbreviations: LV, left ventricle; Ao, aortic; LVOT, left ventricular
outflow tract. Source: From Ref. 5.
procedure is ignored in many laboratories, relying instead on

the pressure recorder’s internal calibration device. If internal
calibration is used, two absolute tenets must be observed. First,
after the catheters (or double lumen catheter) are placed in the
circulation, the closely approximated lumens must record iden-
tical pressures because in that position there can be no gradient.
If the pressures are not identical, (i) ensure that the lines
connecting the catheters to the transducers are properly flushed
and free of kinks, (ii) be sure the zero references of the two
Figure 11.1 Various catheter positions that could be used to obtain
transducers are accurate and identical, and (iii) recalibrate both
the transvalvular gradient in aortic stenosis. 1, left ventricular body;
2, left ventricular outflow tract; 3, ascending aorta, coronary level; 4, transducers. If these steps do not mitigate the errors in the
ascending aorta, distal to coronaries; and 5, femoral artery sheath. pressure being recorded, mercury calibration is then necessary.
The most accurate data are obtained when the left ventricle catheter Second, the above process only proves that the catheters are
is placed in position 1 and the aortic catheter is placed in position 4. recording identical pressures, but not necessarily that the
Source: From Ref. 4. pressures are accurate. To be certain that the pressures
recorded invasively are at least in the “ballpark” of accuracy,
their pressure should be similar to cuff pressure measured by
sphygmomanometery.
It should be recognized that some recording devices
gradient. Because it takes a finite period of time for the pulse to
falsely “assume” that the operator is going to use the FA as
reach the FA, it is impossible for LV and FA pressures to be
the distal recording site despite its inaccuracies. The recorder
simultaneous. This requisite misalignment overestimates the
then offsets the distal pressure to compensate for the expected
true pressure gradient. Unfortunately, manual realignment
delay in registering it. The result will be that the upstroke of the
causes substantial underestimation of the gradient. Because
Ao pressure will actually precede that of the LV pressure, an
invasive hemodynamics are only recorded when the diagnosis
obviously physiologic impossibility leading to inaccurate
is in doubt, it is crucial to obtain the best data possible. This is
pressure recording.
done by obtaining pressures from two properly placed lumens
(or transducers if micromanometer catheters are used). To place
the lumens properly, either two separate catheters can be Accurate Cardiac Output Determination
employed or a double lumen catheter is used. In either case, The second datum crucial to the accurate assessment of stenosis
scrupulous attention to proper damping and catheter flushing severity is the CO. The gold standard for this determination is
are requisite to obtaining an accurate gradient. use of the Fick principle, which states that the oxygen consumed
by the body is the product of O2 delivery (CO) and O2 extraction
Accurate Transducer Calibration by the tissues. Thus, O2 consumption ¼ CO (arterial O2
The most effective way to ensure accurate transducer calibra- venous O2 content, i.e., AO2 VO2D). Rearranging the terms,
tion is to connect the transducer to a mercury manometer CO ¼ O2 consumption/AO2 VO2D. Expanding the latter term,
because pressure is defined as mmHg. The pressure displayed AO2 VO2D ¼ (arterial O2 saturation venous O2 saturation)
from the transducer must be identical to the reading from the hemoglobin (Hb) concentration (g/dL) 1.36 cc O2/g Hb
mercury manometer, or if not, adjusted to be so. Frankly, this 10 dL/L. The principles for the use of the Fick principle are (i) an
[2/7/010/8:5:37] [118–123]
accurate measurement of O2 consumption, (ii) accurate measure- perfected. Their formula was vetted against data from patients
ments of O2 saturations in arterial and venous blood (the latter with mitral stenosis (MS) where they assumed a left ventricular
best taken from the pulmonary artery where mixing is ideal), and end diastolic pressure of 10 mmHg and substituted wedge
(iii) an accurate measurement of hemoglobin concentration. pressure for left atrial pressure (LAP) to obtain the transmitral
Unfortunately, O2 consumption is rarely measured today but gradient. The actual mitral valve area (MVA) was measured
rather is assumed from standard tables usually estimated as from autopsy or surgical specimens and the data compared
function of body size and habitus in normal subjects. However, with their calculations. They then substituted an empirical
the patient with severe valve disease is hardly normal and constant (C) for Cv and Cc. The empirical constant simply
significant error in CO determination and hence valve area reduced the calculated valve area to a value closer to the actual
determination arise when O2 consumption is estimated rather measured valve area. However, the Gorlins had no data for
than measured by actual analysis of expired air (7). This error can developing a similar empiric constant for the Ao valve so they
be as much as 50%. simply assumed a constant of 1.0 warning that when data
By far, the most commonly used method to determine became available the constant should be calculated, but to
CO is thermodilution. this day it never has been.
V1 ðTB TI ÞðSI CI =SB CB Þ60ðsec=minÞ

COTD ¼ Ð1 (1)
0 DTB ðtÞdt Flow Dependence of Calculated Aortic Valve Area
It has been well demonstrated that calculated Ao valve area
where TB is the temperature of blood before injection, TI is the
may be quite flow dependent at COs of <5.0 L/min, with AVA
temperature of the injected saline, SI and SB are the specific
varying directly with flow (8). Increased AVA with increasing
gravities of the injectate and blood, respectively, CI and CB are
flow may be real or factitious. It could be that low CO produced
the
R 1 specific heats of the injectate and blood, respectively, and by a weakened LV is unable to fully open a moderately but not
0 TB ðtÞdt is the area under the time-temperature curve severely stenotic Ao valve. As output is increased by exercise or
recorded by the thermister at the end of the thermodilution
infusion of positive inotropic drugs, the valve is opened more
catheter. When cold saline is injected into the pulmonary artery
widely and output increases more than gradient causing an
the saline mixes with the blood and cools it. The larger the
increase in calculated AVA that truly reflects increased orifice
blood pool and the faster the blood flow is, the smaller will be
area. Conversely, it may be that problems with the Gorlin
the change in downstream blood temperature, thus the area
Formula cause calculated AVA to be flow dependent, either
under the time-temperature curve will be smaller causing the
because it assumes constant flow or because the discharge
calculated CO to be larger than when the blood volume and
coefficients were never calculated. Existing data supports
rate of flow are less. In general, thermodilution is accurate, but
both viewpoints. One recommended method for circumventing
pitfalls exist. The tacit assumption is that all of the coldness of
this problem is to calculate AVA at increasing outputs, con-
the injectate is transferred to the blood stream. However, when
structing a relationship to project what the AVA would be at an
CO is low (an obvious consequence of severe AS), the right
output of 5.0 L/min, where AVA becomes less flow dependent
atrium and ventricle absorb some of the cold, warming up the
(8). Flow dependence of calculated AVA is rarely a problem
blood so that the time-temperature curve area will be factiously
when mean gradient exceeds 40 mmHg since in such cases
small. Other causes of inaccuracy include faulty thermisters,
AVA almost always falls in the severe range of AS irrespective
tricuspid regurgitation, and intracardiac shunts.
of CO.
The Gorlin Formula

In 1951, the Gorlins published their formula for calculating Assessment of Inotropic Reserve
valve area (A). The outcome of aortic valve replacement (AVR) for patients
with AS is usually excellent, resulting in substantial prolonga-
F F F
A¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffi (2) tion of life, a dramatic improvement in symptoms, and an
Cv 2gh Cc Cv Cc 2 980 h ðCÞð44:3Þ h improvement in LV function (9). Even when preoperative ejec-
tion is markedly reduced, ejection fraction (EF) may return to
where F, flow; Cv, the constant of velocity dissipation; Cc, normal following removal of a large transvalvular gradient and
constant of orifice contraction; g, acceleration due to gravity, the afterload that accompanies such a gradient (10). However,
and h, transvalvular pressure gradient (2). Cv accounts for the this is not the case in patients with a low gradient (<30 mmHg
loss of energy as blood flows through the valve since not all of mean gradient) and low EF (<0.30) (11,12). Such patients have
the driving gradient is converted to flow. Cc accounts for the severe myocardial dysfunction and a poor prognosis. However,
fact that as blood flows through an orifice it tends to stream some such patients may improve substantially following AVR.
through the middle so that the physiologic aperture is less than Many are patients that have inotropic reserve (Fig. 11.3) (13). If
the anatomic one. Because flow only occurs when the valve is stroke volume increases by 20% or more during the infusion of
open during each beat, flow ¼ CO/heart rate flow duration. dobutamine and gradient increases concomitantly, prognosis
Aortic valve area (AVA) ¼ following AVR is acceptable. For patients lacking inotropic
CO=ðSEPÞðHRÞ reserve, operative risk is as high as 30% although even then
A¼ pffiffiffiffiffiffiffi (3) some patients still improve if they survive AVR. A third group
44:3 C DP
of patients exists where stroke volume increases but gradient
where SEP is systolic ejection period. does not, resulting in a large increase in calculated AVA (14). It
When the Gorlins devised their formula it was consid- is thought that such patients have moderate but not severe AS
ered malpractice to cross the Ao valve retrogradely as it is done and primary muscle disease. When increased output is pushed
today and other methods for entering the LV had not yet been through such a valve it opens more widely (see above). In such
[2/7/010/8:5:37] [118–123]
position by withdrawing highly oxygen-saturated left atrial

blood from the catheter while it is wedged. When comparison
of PCWP with direct measurement of LAP is made using this
technique the difference is usually less than 3 mmHg, while
comparisons made in the absence of oxymetric confirmation
show larger overestimation of LAP by PCWP (16).
Exact calibration and zeroing of the transducers is crucial
in establishing the transmitral gradient. Whereas an error of 3
to 4 mmHg would only rarely affect clinical decisions in AS,
this magnitude of gradient could be one-third of the total in MS
and lead to serious misdiagnosis.
The Gorlin Formula in MS

The Gorlin formula applied to the mitral valve is

CO=ðHRÞðDFPÞ 2
DP ¼ (4)
ðMVAÞð44:3Þð0:85Þ
Figure 11.3 The outcome of aortic stenosis patients with low
gradient and low ejection fraction according to inotropic reserve where DFP is the diastolic filling period and 0.85 is the empiric
and therapy. Group I patients had inotropic reserve and a satisfac- constant derived by the Gorlins. MVA calculated by the Gorlin
tory outcome with AVR—far better than with medical therapy. Group formula is generally less flow dependent than calculated AVA,
II failed to have inotropic reserve and had a very high operative risk, and calculated MVA usually compares well to actual planime-
although they still did better with AVR than with medical therapy.
tered measurements of MVA (17). Why there is better agree-
Abbreviation: AVR, aortic valve replacement. Source: From Ref. 13.
ment with invasive data used to calculate MVA and
noninvasive techniques than for AVA is uncertain. Perhaps it
is due to employment of the empiric constant or perhaps the
cases (termed pseudo-AS) (15), it is believed, but not certain, MVA changes less than does AVA with flow or perhaps it is a
that AVR would not be of benefit since it is myocardial disease combination of both phenomena.
rather than valve disease which lies at the crux of the problem.
While inotropic reserve is often tested for in the echo- Shortcut for Calculating AVA
cardiography laboratory, the catheterization laboratory is also and MVA: the Hakki Formula
ideal for making such determinations since careful hemody- Because the heart rate, filling time, and square root of the
namic measurements can be made before and during the acceleration due to gravity typically equal a similar constant
inotropic challenge. Further, one cause of failed inotropic in most patients, a rough approximation of valve area can be
reserve can be the presence of obstructive coronary artery made simply by dividing the CO (in liters) by the square root of
disease where increasing intropy can induce ischemia leading the gradient (18). This shortcut is useful in making a quick
to misinterpretation of the test results. In the catheterization ballpark estimation of valve stenosis severity in the catheter-
laboratory, coronary arteriography can be performed to estab- ization laboratory before off-line calculations can be employed.
lish the presence or absence of coronary disease allowing for
better understanding of the results of the inotropic challenge.
TRICUSPID STENOSIS
Tricuspid stenosis (TS) is rare today in developed nations
MITRAL STENOSIS because its major cause, rheumatic heart disease, is also rare.
The hemodynamic determination of MVA in MS employs the
TS may also result from carcinoid syndrome. The diagnosis is
same principles as the hemodynamic determination of AS
usually made echocardiographically and its clinical significance
severity. It requires an accurate transmitral valvular gradient,
is usually well defined at the bedside by examination of the
an accurate CO, and an accurate formula relating the two.
neck veins from which central venous pressure can be esti-
mated. If the diagnosis is uncertain, invasive hemodynamics
Accurate Pressure Gradient may be helpful. A double lumen catheter (or separate right
In common practice, the transmitral gradient is obtained using atrial and right ventricular catheters) is advanced across the
direct measurement of LV pressure, while pulmonary capillary tricuspid valve so that one lumen is positioned on either side of
wedge pressure (PCWP) is used as a surrogate for LAP. While the valve and the gradient is recorded. Since there is no agreed
PCWP may overestimate LAP and therefore overestimate the on valve area that constitutes severe TS, severity is estimated
gradient and the severity of MS, overestimation is minimized to from the gradient alone. A mean gradient of >5 mmHg is
a few millimeters of mercury when careful technique is fol- clinically important as it will yield a systemic venous pressure
lowed (16). In most labs, PCWP is obtained using a Swan-Ganz >10 mmHg at rest, usually enough venous hypertension to
balloon-tipped catheter. Once the balloon is inflated, the cath- cause symptoms.
eter is advanced to the wedge position at which time the
pressure wave form changes from that of a pulmonary artery
tracing to that of the PCWP. However, the change in wave form PULMONARY STENOSIS
does not by itself guarantee that the catheter is fully wedged The severity of pulmonary stenosis is inferred from the mag-
and that an accurate LAP is being measured. To ensure that the nitude of the transvalvular gradient. Mean gradients of less
catheter is truly wedged, it is necessary to confirm the wedge than 25 mmHg are considered hemodynamically insignificant
[2/7/010/8:5:37] [118–123]
while gradients in excess of 75 mmHg indicate severe disease MR (20). In fact, the presence of a large v wave is neither sensitive
requiring intervention. Mean gradients between 25 and 75 nor specific for the presence of severe MR. The v wave height is
mmHg are intermediate with a decision to intervene based on related to LA systolic volume and compliance. A large regurgi-
symptomatology (19). tant volume (RV) and normal or low LA compliance will yield a
large v wave. However, when severe MR coexists with a large
VALVULAR REGURGITATION compliant LA, the v wave may be of normal amplitude. On the
Both the invasive and noninvasive assessment of lesion severity other hand, large v waves may occur in acute heart failure
for regurgitant lesions tends to be less accurate than is the without MR and also with ventricular septal defects.
assessment of valvular stenosis. While relatively precise meth-
ods for regurgitation severity are available, they are more Angiography
difficult to apply than are the quantitative approaches to Aortic Regurgitation
stenosis severity. Thus, uncertainty about lesion severity fol- Aortography has the potential to add significant data in the
lowing noninvasive evaluation is not uncommon for Ao and evaluation of the patient with AR. Whereas Doppler interroga-
mitral regurgitation (MR), in turn necessitating invasive eval- tion of the valve images only the velocity of the regurgitant jet,
uation. In the catheterization laboratory, evaluation of pres- injection of contrast medium into the aorta during aortography
sures and wave forms at rest and with exercise, quantification visualizes actual flow of opacified blood from the aorta across
of regurgitant flow and regurgitant fraction (RF), and visual- the leaking valve into the LV. The severity of AR is classified on
ization of the regurgitant flow contribute to the assessment of the basis of the density of this opacification. Mild AR (1+) is
regurgitant severity. diagnosed when the regurgitant dye fails to opacify the whole
LV. Moderate AR (2+) is thought present when contrast faintly
Wave Forms opacifies the entire LV cavity. Moderately severe AR (3+) is
Aortic Regurgitation present when opacification of the LV is equal to that of the
Figure 11.4 shows some of the classic features of the hemody- aorta, and severe AR (4+) is present when opacification of the
namic of aortic regurgitation (AR) (15). FA systolic pressure is LV exceeds that of the aorta. In performing aortography, it is
about 50 mmHg greater than LV systolic pressure (Hill’s sign). important to inject enough contrast to fully opacify the aorta
The mechanism of this phenomenon is not entirely clear. It is and the enlarged LV, usually 60 cc of contrast injected over
postulated that standing waves along the Ao periphery sum three seconds.
with the systolic pressure wave produced by the increased total
stroke volume of AR to augment Ao systolic pressure as it Mitral Regurgitation
moves down the aorta to the FA. Ventriculography has a similar advantage in visualizing MR as
Ao diastolic pressure and LV diastolic pressure equalize aortography does in AR; it images actual flow instead of flow
(diastasis) because the physical barrier between the aorta and velocity. Severity is judged by the level of LA opacification
LV is largely removed. In turn there is a rapid rise in LV during injection of contrast into the LV. The scale used for MR
diastolic pressure because the LV is filling from both the LA is similar to that for AR, grading the density of dye in the LA
and from the aorta. In chronic AR there is usually a very wide instead of in the LV. However, the tendency of ventriculography
pulse pressure indicative of the large total stroke volume to cause ventricular ectopy (which by itself causes MR) has made
ejected into the aorta. the use of ventriculography uncommon in assessing MR sever-
ity. However, a diagnostic ventriculogram may be obtained by
Mitral Regurgitation placing the injection catheter just underneath the mitral valve
Much has been written about the implications of a large v wave in and using a test injection to evaluate the potential for ectopy. If
the LA (or PCWP) pressure tracing with regard to the severity of the test is unsatisfactory the catheter is repositioned and the
injection repeated. As with AR, enough contrast must be injected
to opacify the two enlarged chambers of interest.
Volumetric Quantification
More precise assessment of the amount of AR or MR can be
made by measuring the RV, which is the difference between
total and forward stroke volume. Total stroke volume (SVt) is
all that is ejected from the LV and can be calculated as end
diastolic minus end systolic volume, where the volumes are
determined angiographically (see chap. 21). Forward stroke
volume (SVf) is determined as CO from the Fick or thermodi-
lution methods divided by the heart rate. The volume regur-
gitated back into the receiving chamber (the LV in AR and the
LA in MR), RV, is the difference between total and forward
stroke volume (SVt SVf). RF ¼ RV/SVt. A RF > 0.50 or
regurgitant flow >60 cc is considered to be severe AR or MR (3).
Figure 11.4 Simultaneous recording of left ventricular and right
femoral artery pressures in a patient with severe aortic regurgitation. Exercise Testing
Abbreviations: LV, left ventricle; RFA, right femoral artery. Source: In most cases, symptoms develop during exercise, yet hemody-
From Ref. 21. namics are measured at rest. Therefore, exercise testing during
hemodynamic recording can be very illuminating. Either
[2/7/010/8:5:37] [118–123]
supine bicycle exercise or isometric handgrip exercise can be Society of Cardiovascular Anesthesiologists: endorsed by the
employed. Handgrip exercise is well suited for evaluation of Society for Cardiovascular Angiography and Interventions and
valvular regurgitation because it increases afterload by raising the Society of Thoracic Surgeons. Circulation 2006; 114:e84–e231.
mean arterial blood pressure by 10 to 20 mmHg (21). Normal 4. Assey ME, Zile MR, Usher BW, et al. Effect of catheter positioning
on the variability of measured gradient in aortic stenosis. Cathet
subjects and patients with well-compensated AR or MR main-
Cardiovasc Diagn 1993; 30:287–292.
tain normal LV filling pressure and increase CO with handgrip. 5. Pasipoularides A. Clinical assessment of ventricular ejection
On the other hand, in poorly compensated patients filling dynamics with and without outflow obstruction. J Am Coll
pressure may increase dramatically giving insight into the Cardiol 1990; 15:859–882.
mechanism of the patient’s symptoms and supporting the 6. Folland ED, Parisi AF, Carbone C. Is peripheral arterial pressure a
need for intervention. satisfactory substitute for ascending aortic pressure when mea-
suring aortic valve gradients? J Am Coll Cardiol 1984; 4:1207–1212.
Indications for Surgery 7. Hillis LD, Firth BG, Winniford MD. Analysis of factors affecting
the variability of Fick versus indicator dilution measurements of
Surgery is recommended either when symptoms referable to
cardiac output. Am J Cardiol 1985; 56:764–768.
severe valve disease or when there is evidence of LV dysfunc- 8. Blais C, Burwash IG, Mundigler G, et al. Projected valve area at
tion (3) in the case of the AR and MR. Symptomatology is based normal flow rate improves the assessment of stenosis severity in
on obtaining a good history. However, if the source of the patients with low-flow, low-gradient aortic stenosis: the multi-
dyspnea, the commonest symptom of VHD, is unclear, invasive center TOPAS (Truly or Pseudo-Severe Aortic Stenosis) study.
hemodynamics can be helpful in establishing cause. If the Circulation 2006; 113:711–721.
filling pressures are normal both at rest and during exercise it 9. Schwarz F, Baumann P, Manthey J, et al. The effect of aortic valve
is likely that the dyspnea is coming from a noncardiac source. If replacement on survival. Circulation 1982; 66:1105–1110.
the patient has both lung and heart disease, examining PCWP 10. Smith N, McAnulty JH, Rahimtoola SH. Severe aortic stenosis
with impaired left ventricular function and clinical heart failure:
and pulmonary artery pressure (PAP) at rest and exercise can
results of valve replacement. Circulation 1978; 58:255–264.
be diagnostic. A normal PCWP and high PAP suggests severe
11. Carabello BA, Green LH, Grossman W, et al. Hemodynamic
lung disease with increased pulmonary vascular resistance, a determinants of prognosis of aortic valve replacement in critical
condition unlikely to be improved by valve surgery. On the aortic stenosis and advanced congestive heart failure. Circulation
other hand, a high PCWP indicates LV failure as the culprit. 1980; 62:42–48.
LV dysfunction is inferred from the LVEF at ventriculog- 12. Connolly HM, Oh JK, Schaff HV, et al. Severe aortic stenosis with
raphy. An EF of <0.50 for AR and <0.60 for MR is considered low transvalvular gradient and severe left ventricular dysfunc-
evidence of LV dysfunction and an indication for valve replace- tion: result of aortic valve replacement in 52 patients. Circulation
ment or repair (3). 2000; 101:1940–1946.
13. Monin JL, Quere JP, Monchi M, et al. Low-gradient aortic stenosis:
operative risk stratification and predictors for long-term outcome:
SUMMARY a multicenter study using dobutamine stress hemodynamics. Cir-
The diagnosis of valvular heart disease is usually made at the culation 2003; 108:319–324.
bedside and is confirmed by echocardiography. However, in 14. Nishimura RA, Grantham JA, Connolly HM, et al. Low-output,
some cases the severity or disease is still in doubt following low-gradient aortic stenosis in patients with depressed left ven-
noninvasive testing. In such cases careful hemodynamic eval- tricular systolic function: the clinical utility of the dobutamine
uation at rest and/or with exercise can be diagnostic. Further, challenge in the catheterization laboratory. Circulation 2002;
106:809–813.
invasively based angiography can contribute diagnostic infor-
15. Carabello BA, Ballard WL, Gazes PC. Cardiology Pearls. Phila-
mation especially for the regurgitant lesions. delphia: Hanley & Belfus, 1994.
Our modern understanding of cardiac function is based 16. Lange RA, Moore DM Jr, Cigarroa RG, et al. Use of pulmonary
on hemodynamic information obtained invasively from pres- capillary wedge pressure to assess severity of mitral stenosis: Is
sure and flow measurement. These principles still have an true left atrial pressure needed in this condition? J Am Coll
important diagnostic role to play even in the 21st century, but Cardiol 1989; 13:825–831.
for them to be useful, the careful techniques of 60 years ago 17. Wang A, Ryan T, Kisslo KB, et al. Assessing the severity of mitral
must still be employed. stenosis: variability between noninvasive and invasive measure-
ments in patients with symptomatic mitral valve stenosis. Am
Heart J 1999; 138:777–784.
REFERENCES 18. Hakki AH, Iskandrian AS, Bemis CE, et al. A simplified valve
1. Forssmann W. Uber Kontrastdarstellung der Hohlen des lebenr- formula for the calculation of stenotic cardiac valve areas. Circu-
den rechten Herzens und der Lungenschlagader. Munchen Med lation 1981; 63:1050–1055.
Wochenschr 1931; 1974:489–492. 19. Johnson LW, Grossman W, Dalen JE, et al. Pulmonic stenosis in
2. Gorlin R, Gorlin SG. Hydraulic formula for calculation of the area the adult. Long-term follow-up results. N Engl J Med 1972;
of the stenotic mitral valve, other cardiac valves, and central 287:1159–1163.
circulatory shunts. I. Am Heart J 1951; 41:1–29. 20. Fuchs RM, Heuser RR, Yin FC, et al. Limitations of pulmonary
3. Bonow RO, Carabello BA, Kanu C, et al. ACC/AHA 2006 guide- wedge V waves in diagnosing mitral regurgitation. Am J Cardiol
lines for the management of patients with valvular heart disease: a 1982; 49:849–854.
report of the American College of Cardiology/American Heart 21. Grossman W. Stress testing during cardiac catherization: exercise
Association Task Force on Practice Guidelines (writing committee and pacing tachycardia. In: Baim DS, ed. Grossman’s Cardiac Cath-
to revise the 1998 Guidelines for the Management of Patients With eterization, Angiography, and Intervention. 7th ed. Philadelphia:
Valvular Heart Disease): developed in collaboration with the Lippincott, Williams & Wilkins, 2006:283–303.
[2/7/010/8:8:10] [124–135]
12
Hemodynamic assessment for restriction, constriction,

hypertrophic cardiomyopathy, and cardiac tamponade
Paul Sorajja
INTRODUCTION the pressures between the left ventricle and right-sided chambers,
Historically, cardiac catheterization has been the principal diag- which is needed in patients undergoing evaluation for pericardial
nostic modality for the evaluation of the patient with constrictive disease, restrictive cardiomyopathy, or diastolic dysfunction:
pericarditis, cardiac tamponade, and the different forms of l Following vascular access, the right heart catheter is used
cardiomyopathy. The advent of newer cardiac imaging modal-
to acquire oxygen saturations from the inferior and supe-
ities, such as echocardiography, magnetic resonance imaging,
rior vena cavae to calculate mixed venous oxygen content
and computed tomography, has led to a shift in the evaluation of
and to screen for intracardiac shunts.
patients with these disorders, with the initial diagnostic l The right atrial pressure is measured with the catheter in
approach consisting of comprehensive noninvasive imaging fol-
the mid-portion of the right atrium and turned to the
lowing the history and physical examination. The hemodynamic
lateral wall to avoid prolapse across the tricuspid valve.
consequences of these disorders may be accurately delineated in l Ascending aortic pressure is then measured by placement
most instances by Doppler echocardiography. Therefore, in most
of a left heart catheter, followed by advancement of the
cases, the diagnosis and hemodynamic assessment of these
catheter into the left ventricle to measure simultaneous left
entities can be made without the need for cardiac catheterization.
ventricular and right atrial pressures.
When clinical and noninvasive findings are discrepant, l The right heart catheter is then advanced into the right
however, an invasive hemodynamic catheterization should be
ventricle to measure simultaneous right ventricular and
considered. Although noninvasive modalities can diagnose the
left ventricular pressures.
presence and severity of these entities, their accuracy is limited. l The right heart catheter then is placed into the pulmonary
Assessment of absolute intracardiac pressures, for example,
artery (PA). Simultaneous saturations from the PA and left
cannot be assessed noninvasively, and cardiac catheterization
ventricle are obtained for calculation of cardiac output by
needs to be performed when this information is needed. Patients
the Fick method.
with complex disorders are increasingly evaluated in the cath- l Finally, the PA wedge pressure is obtained with position
eterization laboratory, whereas patients with straightforward
confirmation by oxygen saturation. Pulmonary arteriolar resis-
conditions are assessed primarily with noninvasive methods.
tance (PAR) may be calculated from these measurements.
Thus, obtaining accurate and clinically relevant data from inva-
sive cardiac catheterization has become increasingly important. For patients with hypertrophic cardiomyopathy (HCM),
transseptal catheterization should be considered in the evalu-
ation of left ventricular outflow tract (LVOT) obstruction. The
ANATOMIC CONSIDERATIONS advantages of transseptal catheterization over a retroaortic
The approach to hemodynamic assessment of patients with approach are discussed separately in the section on HCM.
constrictive pericarditis, cardiac tamponade, and the different While all patients should be fasting before the catheter-
forms of cardiomyopathy should be individualized. Proper ization procedure, intravenous fluids should be administered
planning of the procedure requires full knowledge of what to patients who have a long waiting period between their last
data are known, what clinically relevant information is oral intake and the procedure. This prevents the hemodynamic
required, and a comprehensive differential diagnosis of the measurements from being taken during a low-output, low-
patient’s problems. The vascular access sites and approach to volume state. Patients can be lightly sedated, but should be
gathering data should be delineated fully before proceeding. awake to simulate the hemodynamic milieu of their outpatient
Where both right and left heart catheterization is needed, state with close approximation of the heart rate and blood
a standard approach begins with obtaining arterial and venous pressure that occurs in their usual daily activities. No paren-
access. The femoral sites can be utilized in most situations; teral oxygen should be administered prior to the procedure to
however, the internal jugular approach will facilitate the per- allow measurements of oxygen saturations.
formance of the right heart catheterization in cases involving
severe tricuspid regurgitation with enlarged right-sided cham-
bers. The internal jugular approach should also be favored
FUNDAMENTALS
Key fundamentals of hemodynamic assessment include the
when an endomyocardial biopsy is required (e.g., suspicion
following:
of infiltrative cardiomyopathy), and if hemodynamic informa-
tion during supine bicycle exercise is desired. l A comprehensive differential diagnosis of the patient’s
The following sequence of catheter placement and clinical problems aids in the planning of the invasive
advancement allows prospective examination of the relation of hemodynamic evaluation.
[2/7/010/8:8:10] [124–135]
HEMODYNAMIC CATHETERIZATION 125
l LVOT obstruction in HCM is dynamic and highly depen- l Endomyocardial biopsy has low clinical yield and does not
dent on ventricular load and the contractile state. Physical significantly impact therapy in most cardiac disease states.
or pharmacological provocation should be performed to However, endomyocardial biopsy should be considered in
determine the presence of latent obstruction in symptom- patients with possible acute fulminant giant cell myocardi-
atic patients with no evidence of a significant LVOT tis, where immunosuppressive therapy may be beneficial.
gradient at rest. l Accurate determination of PAR and its reversibility is an
l In patients with obstructive HCM, the typical response on essential component of the comprehensive invasive hemo-
the post–premature ventricular contraction (PVC) beat is a dynamic evaluation.
decrease in the pulse pressure and an increase in the LVOT l With few exceptions, pericardiocentesis is best performed
gradient (Brockenbrough sign). In patients with fixed aortic under echocardiographic guidance.
stenosis, the pulse pressure increases on the post-PVC beat.
l Transseptal catheterization provides the most complete
and accurate hemodynamic information in patients with INDICATIONS
HCM. A retroaortic approach can also be used with under- Table 12.1 lists indications for use of hemodynamic assessment
standing of the potential pitfalls. for specific patient conditions. Indications for patients with
l Invasive assessment of restrictive cardiomyopathy and suspected HCM, restrictive cardiomyopathy, and constrictive
constrictive pericarditis should entail the use of high- pericarditis are discussed below.
fidelity, micromanometer tip catheters to avoid artifacts
due to overdamping.
l Early rapid ventricular filling (i.e., dip and plateau pattern) Hypertrophic Cardiomyopathy: Indications
can be seen in patients with restrictive cardiomyopathy, For the majority of patients with HCM, two-dimensional and
constrictive pericarditis, or any volume overload state that Doppler echocardiography reliably assesses the presence and
results in a decrease in effective myocardial or pericardial severity of LVOT obstruction (1). However, in some circum-
compliance. stances echocardiography may be inaccurate. Mitral regurgita-
l Both traditional and respiratory criteria should be utilized tion frequently accompanies LVOT obstruction in HCM. The
to distinguish restrictive cardiomyopathy from constrictive presence of mitral regurgitation may contaminate the LVOT
pericarditis. Respiratory criteria are (i) dissociation of intra- signal, resulting in erroneous or inaccurate estimates of the
thoracic and intracavitary pressures and (ii) discordance of LVOT gradient (Fig. 12.1). The velocity jet of mitral regurgita-
the right and left ventricular systolic pressures. tion that occurs secondary to LVOT obstruction also is eccen-
l The hemodynamic hallmarks of cardiac tamponade are tric, and thus is difficult to quantify with current noninvasive
pulsus paradoxus and loss of the y descent in the atrial methods. In some patients, a high intracavitary velocity can be
waveform. mistaken on Doppler echocardiography for an LVOT gradient
Table 12.1 Indications for Use of Hemodynamic Assessment for Specific Patient Conditions
Condition Indication
Hypertrophic cardiomyopathy Assess LVOT obstruction; assess severity of mitral regurgitation; evaluate diastolic function;
perform alcohol septal ablation.
Restrictive cardiomyopathy Confirm diagnosis and rule out other potential causes of heart failure (e.g., constrictive pericarditis,
cor pulmonale); measure pulmonary arteriolar resistance for potential transplant candidates;
perform endomyocardial biopsy.
Constrictive pericarditis Establish the diagnosis and examine for other potential causes of heart failure.
Cardiac tamponade Invasive assessment is not necessary for the diagnosis of tamponade, but the typical hemodynamic
findings of tamponade during cardiac catheterization need to be understood.
Figure 12.1 Contamination of the Doppler left ven-

tricular outflow tract signal with mitral regurgitation in
a patient with obstructive hypertrophic cardiomyop-
athy. Contamination from mitral regurgitation leads to
an erroneous calculation of the left ventricular outflow
tract gradient (left). The correct Doppler left ventric-
ular outflow tract gradient (right). Cardiac catheter-
ization is indicated when it is difficult to separate the
velocities and the severity of left ventricular outflow
tract obstruction is unclear. Abbreviations: Ao,
ascending aortic pressure; LV, left ventricular pres-
sure.
[2/7/010/8:8:10] [124–135]
Figure 12.2 Restrictive cardiomyopathy. Gross

specimen showing severe biatrial enlargement
due to restrictive cardiomyopathy (left). Biatrial
enlargement is easily seen on echocardiography
(apical four-chamber transthoracic view) (right);
arrow shows dilated pulmonary vein that suggests
elevated diastolic filling pressures. Abbreviations:
LA, left atrium; LV, left ventricle; RA, right atrium;
RV, right ventricle. Source: Courtesy of Dr William
D. Edwards, Mayo Clinic, and Ref. 5.
(2). For patients with combined valvular stenosis and subaortic the subaortic pressure and thus should not be used in these
obstruction, cardiac catheterization may be necessary to assess patients.
the relative contributions of these different levels of obstruc- Fluid-filled catheters can reliably measure mean and
tion. For patients with cardiovascular symptoms and no evi- absolute intracardiac pressures. For analysis of pressure wave-
dence of significant LVOT obstruction at rest, cardiac forms, however, instantaneous recordings with high-fidelity
catheterization with provocative maneuvers is utilized to delin- micromanometer tip catheters should be utilized (Millar Instru-
eate the presence of latent LVOT obstruction (3). ments, Houston, Texas, U.S.). These catheters should be cali-
brated to fluid-filled pressures at baseline, and calibration
needs to be repeated following any catheter repositioning. For
Restrictive Cardiomyopathy and Constrictive right heart pressure measurements, a 6 or 7 French (Fr) single
Pericarditis: Indications lumen balloon wedge catheter (Arrow International, Teleflex
The diagnosis of restrictive cardiomyopathy is made by the Medical, Research Triangle Park, North Carolina, U.S.) is rela-
presence of diastolic dysfunction, dilated atria, and the absence tively rigid with a large bore that will accommodate a 2-Fr
of ventricular hypertrophy or dilatation on noninvasive imag- high-fidelity micromanometer tip catheter.
ing (Fig. 12.2) (4). In these patients, systolic function is pre- For patients with irregular heart rates (e.g., atrial fibril-
served or out of proportion to the degree of diastolic lation), temporary pacing should be considered to maintain
dysfunction. For patients presenting with restrictive cardiomy- consistent R-R intervals to improve the diagnostic interpreta-
opathy, the major differential diagnosis that must be consid- tion of the hemodynamic findings. If possible, continuous
ered is constrictive pericarditis. If constrictive pericarditis is recording of all hemodynamic pressures should be made to
present, complete pericardiectomy can result in relief of symp- allow retrospective review of these pressures throughout the
toms and improvement of longevity (6). However, for patients entire study.
with idiopathic restrictive cardiomyopathy, there is no curative
treatment apart from cardiac transplantation (7). During car-
diac catheterization, it is also important to examine for other CLINICAL ASPECTS
potential causes of heart failure. The presence of severe intrinsic Hypertrophic Cardiomyopathy
pulmonary hypertension, tricuspid regurgitation, and left ven- HCM is a heritable cardiac disorder with a prevalence of 1 in
tricular or right ventricular failure will lead to symptoms and 500 persons in the general population. It is defined, according
signs similar to those of constrictive pericarditis and restrictive to the World Health Organization, by the presence of severe
cardiomyopathy. myocardial hypertrophy in the absence of a local or systemic
etiology (8). In the late 1980s, molecular studies demonstrated
that HCM is due to genetic mutations in sarcomeric contractile
EQUIPMENT proteins (9). Several hundred mutations in 10 different genes
Accurate measurement of intracardiac pressures with fluid- have since been identified. The diagnosis of HCM typically is
filled catheters requires the use of rigid, large-bore catheters made by two-dimensional echocardiography. In most instan-
with minimization of the tubing length between the catheter ces, Doppler echocardiography can accurately diagnose and
and pressure transducer. Awareness of potential errors of quantify LVOT obstruction. The late peaking systolic velocity
measurement due to catheter whip, entrapment, damping, can be converted to a pressure gradient using the modified
and other artifacts is always necessary during an invasive Bernoulli equation (gradient ¼ 4velocity2).
hemodynamic study. The use of coronary catheters with single
end holes and potential for entrapment (e.g., right Judkins Pathophysiology
catheter) should be avoided. In patients with intracavitary Diastolic abnormalities are the major pathophysiological mech-
gradients (e.g., HCM and known or suspected LVOT obstruc- anisms contributing to signs and symptoms for patients with
tion), multiple shaft side holes on a left ventriculography cath- HCM. These abnormalities arise from impaired myocardial
eter (i.e., pigtail) also will lead to errors in the measurement of relaxation and poor compliance in the presence of altered
[2/7/010/8:8:10] [124–135]
Figure 12.3 Left ventriculogram from a patient

with severe basal septal hypertrophy due to
hypertrophic cardiomyopathy. End-diastolic
frame showing severe basal septal hypertrophy
(asterisk) (left). End-systolic frame showing hyper-
dynamic systolic function with papillary muscle
hypertrophy (arrowheads) (right). Abbreviation:
LV, left ventricle. Source: From Ref. 11.
Figure 12.4 Left ventriculogram from a patient

with severe apical hypertrophy due to hypertrophic
cardiomyopathy. End-diastolic frame shows the
typical “spade-shaped” configuration seen on ven-
triculography (left). End-systolic frame of the same
patient showing systolic obliteration of the left
ventricular cavity with a small apical pouch
(arrow) (right). Abbreviations: Ao, ascending
aorta; LV, left ventricle. Source: From Ref. 11.
loading conditions, ventricular nonuniformity, myocardial changes in preload, afterload, and the contractile state of the
ischemia, and severe hypertrophy. The cumulative result of left ventricle. Mitral regurgitation may be related to intrinsic
diastolic dysfunction is an increase in left ventricular filling valve abnormalities in a subset of patients, and such abnormal-
pressures, which leads to typical symptoms of dyspnea and ities have implications for septal reduction therapy.
angina. In patients with HCM, there may be a significant During catheterization, HCM should be suspected when
discrepancy between the left atrial pressure and left ventricular there is a small, hypertrophied left ventricle with hyperdy-
end-diastolic pressure. Therefore, measurements of both of namic systolic function on left ventriculography. Left ventricu-
these pressures should be made if possible. lography may demonstrate regional hypertrophy, such as basal
Dynamic LVOT obstruction may be present in up to two- septal hypertrophy (Fig. 12.3). In patients with the apical form
thirds of patients with HCM (3). Because the presence of LVOT of HCM, a typical spade-shaped configuration is on left ven-
obstruction serves as the basis for therapy, it is important to triculography (Fig. 12.4). A dynamic LVOT obstruction can also
document the presence and severity of the LVOT gradient. Two be suspected if there is a gradient between the left ventricular
mechanisms lead to the development of dynamic LVOT apex and base or if a “spike and dome” pattern is on the aortic
obstruction: (i) septal hypertrophy and narrowing of the pressure trace.
LVOT promote the generation of Venturi forces that accelerate
during ventricular emptying and pull the mitral apparatus Catheterization Techniques
anteriorly; and (ii) anterior papillary muscle displacement Measurement of the LVOT gradient should be made with
subjects the mitral leaflets to systolic intraventricular currents simultaneous ascending aortic and left ventricular pressures.
that drag the apparatus anteriorly (3). Decreased mitral leaflet Because of peripheral amplification, femoral tracings should
coaptation occurs because of systolic anterior motion of the not be used. In patients with significant LVOT obstruction, a
mitral valve, leading to mitral regurgitation in patients with typical spike and dome pattern appears in the ascending aortic
LVOT obstruction (10). It is important to note the dynamic pressure (Fig. 12.5). The optimal method for measurement of
nature of LVOT obstruction and secondary mitral regurgita- LVOT obstruction is to measure the left ventricular pressure via
tion, and the marked sensitivity of these abnormalities to a transseptal approach to avoid catheter entrapment. The
[2/7/010/8:8:10] [124–135]
from a retrograde approach and may falsely elevate the LVOT

gradient. If catheter entrapment is present, there will be a lack
of systolic pulsation when the catheter is disconnected and the
left ventricular pressure contour will be dampened. In the
retrograde approach, it is also important to confirm that
the catheter position is beneath the level of LVOT obstruction
(Fig. 12.7). A hand contrast injection should be performed to
ensure that the end of the catheter is floating freely within the
left ventricular cavity.
The dynamic nature of LVOT obstruction due to HCM
will lead to a Brockenborough response. On the beat following
a premature contraction, the LVOT gradient increases and the
pulse pressure decreases (Fig. 12.8). Because of its high sensi-
tivity to loading conditions, variation of the LVOT gradient
may also be seen in different phases of respiration (Fig. 12.9).
If the resting LVOT gradient is <50 mmHg in a patient
with HCM, provocative maneuvers should be performed in the
catheterization laboratory. These maneuvers include the Val-
salva maneuver, amyl nitrate inhalation, and drug provocation.
The optimal drug to use for provoking LVOT obstruction is
isoproterenol because of its b1 and b2 agonist activity (12,14).
Simultaneous echocardiography should be performed during
isoproterenol infusion to determine the location of obstruction
(e.g., systolic anterior motion of the mitral valve, mid-cavitary
obstruction) when a gradient is detected during these maneu-
Figure 12.5 Dynamic left ventricular outflow obstruction in hyper- vers. It is the subset of patients who have both systolic anterior
trophic cardiomyopathy. Arrow indicates typical “spike and dome” motion and LVOT gradient >50 mmHg (either at rest or during
configuration, indicating dynamic systolic obstruction. Abbrevia- provocation) who will benefit from septal reduction therapy (1).
tions: Ao, ascending aorta; LA, left atrium; LV, left ventricle.
Restrictive Cardiomyopathy
Restrictive cardiomyopathy is characterized by a nondilated,
rigid ventricle, resulting in severe diastolic dysfunction and
catheter can be placed in the left ventricular inflow region restrictive filling. Desmin and troponin I mutations have been
immediately distal to the opening of the mitral valve leaflets described in patients with restrictive cardiomyopathy (15–17).
(Fig. 12.6). Using a transseptal sheath with a side-arm, simul- Infiltrative cardiomyopathies, such as amyloidosis, hemochro-
taneous left atrial pressure also can be obtained (Fig. 12.5). matosis, and sarcoidosis, can present with an appearance of a
If transseptal catheterization cannot be performed, left restrictive cardiomyopathy. Endomyocardial biopsy can be
ventricular pressure is obtained by a retrograde approach performed for patients in whom restrictive cardiomyopathy
across the aortic valve. In these instances, a pigtail catheter due to infiltrative disorders is suspected (see section “Endo-
with shaft side holes should not be used. Importantly, catheter myocardial Biopsy” below). Restrictive cardiomyopathy also
entrapment may occur with left ventricular pressure measured may result from radiation therapy and eosinophilic syndromes.
Figure 12.6 Transseptal versus retrograde approach for

assessment of left ventricular outflow tract obstruction in
hypertrophic cardiomyopathy. In the retrograde approach,
the catheter becomes easily entrapped and lead to errone-
ous left ventricular pressure measurements (left). In the
transseptal approach, the left ventricular catheter can be
easily positioned in near the mitral inflow (right). This cath-
eter position leads to the most accurate measurements of
left ventricular pressure in patients with subaortic obstruc-
tion. Source: From Ref. 12.
[2/7/010/8:8:10] [124–135]
Figure 12.7 Importance of catheter position in the retroaortic approach in hypertrophic cardiomyopathy. When a retroaortic approach is
utilized for assessment of LVOT obstruction in hypertrophic cardiomyopathy, it is important confirm that the catheter position is beneath the
level of LVOT obstruction. A catheter in the ascending aorta and a catheter in the left ventricle (A). With the catheter in this position, the LVOT
gradient is either not detected (C) or measured to be approximately 80 mmHg (D). However, the true gradient is obtained by advancing the left
ventricular catheter further into chamber (B), leading to the correct LVOT gradient of 120 mmHg (E). Abbreviations: LVOT, left ventricular
outflow tract; Ao, ascending aorta; LV, left ventricle.
Figure 12.8 Dynamic left ventricular outflow obstruction (Brockenbrough sign) in hypertrophic cardiomyopathy (left) versus a patient with
aortic stenosis (right). On the beat after a PVC, the increased inotropy leads to dynamic outflow tract obstruction. The left ventricular outflow
tract gradient increases and the pulse pressure decreases. Conversely, in a patient with aortic stenosis, the increase in contractility leads to
an increase in stroke volume and pulse pressure. Abbreviations: Ao, ascending aorta; LA, left atrium; LV, left ventricle; RA, right atrium.
[2/7/010/8:8:10] [124–135]
Figure 12.9 Respiratory variation of the LVOT gradient in hyper-

trophic cardiomyopathy. LVOT obstruction in hypertrophic cardio-
myopathy is highly sensitive to ventricular load. This sensitivity is
evident in this patient during quiet respiration. In expiration, the
transmural pressure decreases, leading to a decrease in afterload
and an increase in LVOT obstruction. During inspiration, the drop in
thoracic pressure leads to an increase in transmural pressure and Figure 12.10 Early rapid ventricular filling. In the right atrial tracing,
afterload, leading to a decrease in LVOT obstruction. Abbreviations: there are rapid x and y descents. The y descent of the right atrial
LVOT, left ventricular outflow tract; Ao, ascending aorta; LV, left pressure tracing corresponds to the early rapid filling phase of the
ventricle. ventricular pressure tracing, which demonstrates the typical dip and
plateau pattern (arrow). These hemodynamic tracings were taken
using high-fidelity micromanometer catheters from a patient with
constrictive pericarditis. Abbreviations: LV, left ventricle; RA, right
atrium. Source: From Ref. 5.
Cardiac Catheterization
The major abnormality in restrictive cardiomyopathy is the
presence of significantly elevated filling pressures in all four
cardiac chambers. Early rapid ventricular filling leads to a “dip Constrictive Pericarditis
and plateau” pattern or “square root sign” in the ventricular Constrictive pericarditis results from pericardial inflammation,
pressure curves during early diastole, and rapid x and y fibrosis, and possibly calcification, with subsequent loss of
descents on the atrial pressure curves (Fig. 12.10). Because elasticity. Radiation therapy, cardiac surgery, trauma, and sys-
underdamping from fluid-filled catheters can mimic early temic diseases that affect the pericardium (e.g., connective
rapid ventricular filling, high-fidelity micromanometer cathe- tissue disease, tuberculosis, malignancy) can lead to constric-
ters should be used if early rapid ventricular filling is tive pericarditis.
suspected. In patients with constrictive pericarditis, the noncompli-
Both restrictive cardiomyopathy and constrictive pericar- ant pericardium is rigid, impairs diastolic filling, and prevents
ditis present with evidence of early rapid diastolic filling and the complete transmission of intrathoracic pressure to the
elevation of diastolic pressures that are out of proportion to intracardiac cavities. Ventricular filling rapidly occurs in early
systolic dysfunction. Traditional criteria for differentiation of diastole and terminates abruptly because of pericardial
restrictive cardiomyopathy from constrictive pericarditis have restraint. The diastolic pressures become equalized or nearly
included the following: equalized in all four cardiac chambers. The total intracardiac
volume is fixed by the noncompliant pericardium. Because the
l Left ventricular end-diastolic pressure exceeds right ven- ventricular septum is not affected in constrictive pericarditis,
tricular end-diastolic pressure by 5 mmHg. bulging of the septum toward the left ventricle occurs during
l PA systolic pressure is >50 mmHg. inspiration and returns toward the right ventricle during expi-
l In the right ventricle, end-diastolic pressure is <0.3 of ration, leading to marked enhancement of ventricular interac-
systolic pressure. tion. This ventricular interaction leads to reciprocal changes in
Nonetheless, these traditional criteria have been found to have the filling and emptying of the right and left ventricles (18).
poor specificity, and they cannot be used in isolation to differ-
entiate restrictive cardiomyopathy from constrictive pericardi- Invasive Hemodynamic Criteria
tis. On the other hand, changes in the hemodynamic pressure Invasive evaluation of the patient with suspected constrictive
relationships during respiration have been shown to be useful pericarditis is accomplished with comprehensive, simultaneous
in distinguishing between these two disorders (see section right and left heart catheterization. High-fidelity micromanom-
“Constrictive Pericarditis” below). eter catheters should be utilized to avoid artifacts from
[2/7/010/8:8:10] [124–135]
underdamping. For patients with atrial fibrillation, temporary

pacing should be performed for constant heart rate intervals
during the study.
Criteria for the diagnosis of constrictive pericarditis
include both traditional hemodynamic criteria and dynamic
respiratory criteria (18–21). Respiration affects ventricular fill-
ing in constrictive pericarditis in a manner that is distinct from
restrictive cardiomyopathy.
l In patients with constriction, the inspiratory fall in thoracic
pressure affects the pulmonary wedge pressure, but ven-
tricular pressure is shielded from respiratory pressure
changes by the pericardial scar. By lowering pulmonary
wedge pressure and presumably left atrial pressure, inspi-
ration leads to a decrease in pressure gradient for ventric-
Figure 12.11 Dissociation of intracavitary and intrathoracic pres-
ular filling. Reciprocal changes occur in right ventricular sures in constrictive pericarditis. In patients with constrictive peri-
filling that are mediated by the ventricular septum (not by carditis, inspiration leads to a decrease in ventricular filling by
increased systemic venous return). These findings are decreasing intrathoracic pressure relative to ventricular diastolic
described as dissociation of the intrathoracic and intracavi- pressure. Conversely, during expiration, positive intrathoracic pres-
tary pressures (Fig. 12.11). sure leads to an increase in ventricular filling. These respiratory
l In patients with constriction, the enhancement of ventric- effects can be seen by examining the changes in the pressure
ular interaction leads to discordant right and left ventricular gradient between the PCWP and ventricular early diastolic pressure
pressures. This discordance typically manifests as recipro- (gray). Abbreviations: PCWP, pulmonary capillary wedge pressure;
cal changes in stroke volume, pulse pressure, or peak LV, left ventricle. Source: From Ref. 5.
systolic pressure during respiration (Fig. 12.12).
Both dissociation of intrathoracic and intracavitary pressures
and enhancement of ventricular interaction are not present in
patients with restrictive cardiomyopathy. In these patients,
inspiration lowers the pulmonary wedge and left ventricular states, when acute volume overload leads to pericardial
diastolic pressures equally. Thus, the pressure gradient for restraint. Examples of this phenomenon include right ventric-
ventricular filling is virtually unchanged during respiration. ular dilatation after right ventricular infarction, severe decom-
Because ventricular interaction is not enhanced, the left pensated left heart failure, severe tricuspid insufficiency, and in
ventricular and right ventricular pressures are concordant acute mitral regurgitation secondary to chordal rupture. In
throughout the respiratory cycle in patients with restrictive addition, equalization of diastolic pressures may not be present
cardiomyopathy. in a patient with constrictive pericarditis who has been diu-
Equalization of pressures in all cardiac chambers is com- resed and has low to normal right atrial pressure. In these
monly said to be a major criterion for constrictive pericarditis patients, the cardiac output will be low, and fluid challenge will
but is nonspecific. This equalization also may be present in be necessary to unveil the hemodynamic findings of constric-
patients with restrictive cardiomyopathy or other disease tive pericarditis (22).
Figure 12.12 Enhancement of ventricular interdepen-

dence. In patients with constrictive pericarditis, the total
ventricular volume is fixed by the noncompliant pericar-
dium. Thus, reciprocal respiratory changes in the filling
of each ventricle occur. These changes are described as
discordance in pulse pressure, systolic pressure, or
stroke volume between the right and left ventricles dur-
ing respiration. Abbreviations: RV, right ventricle; LV, left
ventricle. Source: From Ref. 5.
[2/7/010/8:8:10] [124–135]
Figure 12.13 Tamponade versus constrictive pericar-

ditis. In cardiac tamponade, the y descent in the right
atrial pressure tracing becomes blunted because of
impairment of ventricular filling (arrows). The FA pres-
sure tracing also demonstrates pulsus paradoxus seen
on the second and third full systolic tracings. However,
in patients with constrictive pericarditis, there is early
rapid diastolic filling that manifests as a prominent y
descent in the right atrial tracing and as a “dip and
plateau” sign in the left ventricular pressure tracing
(arrow). Abbreviations: RA, right atrium; FA, femoral
artery; LV, left ventricle.
Cardiac Tamponade The changes in mitral inflow are highly sensitive, and can
Cardiac tamponade may result from any disorder that causes a precede changes in cardiac output, blood pressure, and other
pericardial effusion. The pericardium normally contains 15 to echocardiographic evidence of tamponade (26).
50 mL of fluid between its parietal and visceral layers, and the
intrapericardial pressure approximates intrapleural pressure Invasive Hemodynamics
(5 to þ5 cm H2O). Tamponade occurs when intrapericardial Although cardiac catheterization is usually not needed to make
pressure exceeds the intracardiac pressure. The progression in the diagnosis of cardiac tamponade, recognition of the typical
the effect of a pericardial effusion may be acute or gradual, hemodynamic findings has gained renewed importance in the
depending on the amount and rate of fluid accumulation. The era of increasingly complex cardiac interventions. When per-
most common etiology of tamponade is malignancy, of which formed, the right atrial pressure tracing in tamponade will
breast and lung are the most frequent. Other important etiol- demonstrate prominent x descents and blunted or obliterated
ogies are idiopathic or viral pericarditis, aortic dissection with y descents (Fig. 12.13). Preservation of the x descent occurs
disruption of the aortic valve annulus, complications of inva- because systolic ejection decreases the intracardiac volume and
sive cardiac procedures, uremia, tuberculosis, and pericarditis leads to a transient reduction in intrapericardial and right atrial
or myocardial rupture from myocardial infarction. Unusual pressures. During the remainder of the cardiac cycle, elevated
manifestations of cardiac tamponade include the following: intrapericardial pressure impairs ventricular filling and leads
to blunting or obliteration of the y descent. Intrapericardial
l Low-pressure tamponade, which is tamponade without ele-
pressure rises with fluid accumulation and pericardial
vated jugular venous pressure because of the intracardiac
restraint. Venous return becomes impaired once the intraper-
filling pressures are low (23,24). Examples of this manifes-
icardial pressure exceeds the filling pressure of the heart. This
tation are patients with malignancy or tuberculosis that is
impairment leads to a reduction in cardiac output, followed by
complicated by severe dehydration.
increases in pulmonary venous and jugular venous pressures.
l Localized tamponade, which occurs when a loculated peri-
During inspiration, there is a fall in the driving pressure to fill
cardial effusion is tactically located to cause impairment of
the left ventricle, followed by a reduction in ventricular filling,
ventricular filling. An example of this manifestation is in
stroke volume, and consequently the pulse pressure. These
the postoperative setting, where a loculated effusion is
events during inspiration cause the hallmark finding of pulsus
present in the posterior pericardial space adjacent to the
paradoxus in patients with tamponade (27,28).
atria. A posterior effusion may not be seen with trans-
esophageal echocardiography and must be carefully sought
in a postoperative patient with hemodynamic instability.
l Pneumopericardium, which may be caused by gas-forming
SPECIAL ISSUES
bacterial pericarditis following penetrating chest trauma.
Endomyocardial Biopsy
Endomyocardial biopsy provides myocardial tissue for micro-
Cardiac tamponade should be considered when there is a scopic analysis and can be safely performed using either an
compatible history, hypotension, and an elevated jugular internal jugular or femoral venous access. Particular care must
venous pressure or pulsus paradoxus. The chest x-ray (e.g., be taken to target the apical septum, and to avoid sampling
“water bottle heart”) and electrocardiography (e.g., electrical from the right ventricular free wall and the tricuspid valve
alternans, sinus tachycardia) may be helpful. However, echo- (Fig. 12.14). Biplane fluoroscopy and/or simultaneous echocar-
cardiography is the primary test for the diagnosis. Specific signs diography are helpful in preventing this from occurring. Com-
include collapse of the right atrium and right ventricle, ven- plications are infrequent (<1%), but include tricuspid valve
tricular septal shifting with respiration, and enlargement of the injury and tamponade from cardiac perforation (29).
inferior vena cava (25). Respiratory variation in Doppler mitral Before proceeding with endomyocardial biopsy, it is
inflow velocities occurs early in the evolution of tamponade. important to consider both the expected yield of the biopsy
[2/7/010/8:8:10] [124–135]
Figure 12.14 Proper technique for performing endo-

myocardial biopsy of the right ventricle. Images (A)
and (C) are from one patient; images (B) and (D) are
from a second patient. The bioptome is in a similar
position on the AP view in both patients (A) and (B).
However, in the lateral views (bottom images), the
bioptome is safely positioned against the septum only
in the patient on the left hand side (images A and C).
In the patient on the right hand side (images B and D),
the bioptome is directed anteriorly against the right
ventricular free wall. This could result in performation
and cardiac tamponade. Source: From Ref. 5.
procedure and its therapeutic implications. Endomyocardial For patients with elevated PAR, cardiac transplantation has
biopsy can diagnose myocarditis and infiltrative disorders caus- a poor outcome due to the high afterload imposed on the right
ing cardiomyopathy. However, its sensitivity is limited in dis- ventricle. When the PAR is elevated, vasodilators or inhaled nitric
orders with patchy myocardial involvement (e.g., sarcoidosis, oxide should be used to determine the degree of its reversibility.
lymphocytic myocarditis). Moreover, the histological findings, For patients with normal to mildly elevated filling pressures (left
even if positive, may not alter therapy in other disorders. Thus, atrial pressure < 20 mmHg), nitric oxide, epoprostenol, or nitro-
few indications for endomyocardial biopsy exist in patients with prusside may be administered. For patients with elevated filling
left ventricular systolic dysfunction. For infiltrative cardiomyo- pressures, nitroprusside primarily is used. Repeat measurements
pathies, there are usually clinical clues in the history, electrocar- of PA pressure, left atrial pressure or PCWP, and cardiac output
diography, and laboratory testing, such as iron studies, protein should be made after each increase in infusion rate. The endpoint
electrophoresis, and peripheral eosinophilia. In patients present- of the infusion should be either a PAR of <4.0 Wood units or a
ing with acute fulminant myocarditis, endomyocardial biopsy systolic blood pressure <75 mmHg.
may identify giant cell myocarditis, which may respond favor- Several technical considerations when measuring PAR
ably to immunosuppressive therapy (30–32). include the following:
l The measurements of PA pressure and left atrial pressure
(or PCWP) should be near simultaneous and at held end
Pulmonary Arteriolar Resistance expiration.
Patients being considered for cardiac transplantation require l If PCWP is used in place of direct left atrial pressure
cardiac catheterization to ensure suitability for cardiac trans-
measurement, its accuracy must be confirmed. Phasic
plantation. In these patients, an accurate measurement of PAR
respiratory changes in the pressure and oximetric confir-
is critically important in their evaluation. In other patients,
mation by >95% saturation should be evident.
measurement of PAR also can be helpful to determine appro- l Cardiac output is a critical measurement, and should be
priate medical therapy (e.g., vasodilators). Thus, determination
confirmed by at least two methods. The Fick method
of PAR should be a routine component of a comprehensive
requires steady state conditions and measurement of myo-
invasive hemodynamic evaluation.
cardial oxygen consumption. Thermodilution is better for
Calculation of the PAR is made with measurements of
acute changes in cardiac output.
the PA pressure, left atrial pressure, and cardiac output. The
pulmonary capillary wedge pressure (PCWP) approximates
left atrial pressure except in rare circumstances (e.g., cor
triatriatum, pulmonary veno-occlusive disease). Pericardiocentesis
Historically, pericardiocentesis was performed in a blinded or
PAR ðWood unitsÞ ¼ ½mean PA pressure mean LA pressure electrocardiographic-guided fashion, usually from a subxy-
ðor PCWPÞ=cardiac output where LA is left atrium: phoid approach. Although these techniques are still used in
[2/7/010/8:8:10] [124–135]
some situations (e.g., cardiogenic shock), echocardiographic be difficult to remove percutaneously. A true posterior effusion
guidance is strongly preferred because of the high incidence may be difficult to approach from any thoracic window and
of complications (31,33). Care also should be taken to avoid may require surgery.
pericardiocentesis for tamponade that occurs with aortic dis-
section, as the abrupt return of ventricular ejection may extend
the dissection and lead to acute decompensation (34).
REFERENCES
1. Maron BJ, McKenna WJ, Danielson GK, et al. American College of
l Echocardiography is used to determine the most appro- Cardiology/European Society of Cardiology clinical expert con-
priate portal of entry and needle direction into the peri- sensus document on hypertrophic cardiomyopathy. A report of
cardial effusion, typically the window closest to the the American College of Cardiology Foundation Task Force on
Clinical Expert Consensus Documents and the European Society
effusion. This site frequently is apical, but other locations
of Cardiology Committee for Practice Guidelines. J Am Coll
that have been used are axillary, left or right parasternal, Cardiol 2003; 42:1687–1713.
and the subxyphoid window. The needle trajectory is 2. Jaber WA, Nishimura RA, Ommen SR. Not all systolic velocities
transfixed in the operator’s mind. Care should be taken indicate obstruction in hypertrophic cardiomyopathy: a simulta-
to avoid the internal mammary or intercostal arteries by neous Doppler catheterization study. J Am Soc Echocardiogr 2007;
passing superior to the ribs. The entry site can be marked 20:1009 e5–e7.
with an indelible pen. 3. Maron MS, Olivotto I, Zenovich AG, et al. Hypertrophic cardio-
l Local anesthesia is administered. myopathy is predominantly a disease of left ventricular outflow
l Using the predetermined angulation, a Polytef-sheathed tract obstruction. Circulation 2006; 114:2232–2239.
needle is inserted at the entry site and advanced with 4. Benotti JR, Grossman W, Cohn PF. Clinical profile of restrictive
cardiomyopathy. Circulation 1980; 61:1206–1212.
gentle aspiration into the pericardial space. Once fluid is
5. Oh JK, Seward J, Tajik AJ, eds. The Echo Manual. 2nd ed. Phila-
obtained, the needle is advanced slightly further (*2 mm). delphia: Lippincott Williams & Wilkins, 1996:164.
The Polytef sheath then is advanced over the needle, 6. McCaughan BC, Schaff HV, Piehler JM, et al. Early and late results
followed by withdrawal of the needle. The needle should of pericardiectomy for constrictive pericarditis. J Thorac Cardio-
not be readvanced forward into the sheath once it has been vasc Surg 1985; 89:340–350.
removed. 7. Ammash NM, Seward JB, Bailey KR, et al. Clinical profile and
l Agitated saline is injected into the Polytef sheath via a outcome of idiopathic restrictive cardiomyopathy. Circulation
three-way stopcock under echocardiographic monitoring. 2000; 101:2490–2496.
If contrast does not opacify the pericardial space, then the 8. Richardson P, McKenna W, Bristow M, et al. Report of the 1995
catheter should be repositioned by withdrawal or another World Health Organization/International Society and Federation
of Cardiology Task Force on the Definition and Classification of
needle passage. As previously noted, the needle should not
cardiomyopathies. Circulation 1996; 93:841–842.
be advanced back into the sheath once it has been 9. Geisterfer-Lowrance AA, Kass S, Tanigawa G, et al. A molecular basis
removed. for familial hypertrophic cardiomyopathy: a beta cardiac myosin
l Once the intrapericardial position of the Polytef sheath is heavy chain gene missense mutation. Cell 1990; 62:999–1006.
confirmed, it is exchanged over a guidewire for a 5- to 6-Fr 10. Levine RA, Vlahakes GJ, Lefebvre X, et al. Papillary muscle
introducer sheath followed by placement of a pigtail cath- displacement causes systolic anterior motion of the mitral valve.
eter the pericardial space. The introducer sheath is sub- Experimental validation and insights into the mechanism of
sequently removed to leave only the smooth walled pigtail subaortic obstruction. Circulation 1995; 91:1189–1195.
catheter in place. If needed, reconfirmation of the catheter 11. Sorajja P, Nishimura RA. Myocardial and pericardial disease
location and measurement of intrapericardial pressure can assessment and management, CathSAP-3. American College of
Cardiology Foundation.
be performed with saline injection.
12. Elesber A, Nishimura RA, Rihal CS, et al. Utility of isoproterenol
l The pericardial effusion is removed using either manual to provoke outflow tract gradients in patients with hypertrophic
techniques and/or vacuum bottle. Fluid removal is moni- cardiomyopathy. Am J Cardiol 2008; 101:516–520.
tored with echocardiography. If drainage stops despite 13. Sorajja P, Nishimura RA. The assessment and therapy of valvular
residual effusion on echocardiography, the pigtail is repo- heart disease in the cardiac catheterization laboratory. In: Willer-
sitioned. son JP, Cohn JN, Wellens HH, et al., eds. Cardiovascular Medi-
l The pigtail catheter is aspirated and flushed with hepari- cine. 3rd ed. London: Springer, 2007:463–486.
nized saline every four to six hours. Reapposition of the 14. Vaglio JC Jr, Ommen SR, Nishimura RA, et al. Clinical character-
parietal and visceral pericardial surfaces in this manner istics and outcomes of patients with hypertrophic cardiomyop-
promotes adhesions that prevent fluid recurrence. The athy with latent obstruction. Am Heart J 2008; 156:342–347.
15. Mogensen J, Kubo T, Duque M, et al. Idiopathic restrictive
catheter is removed when the drainage is minimal (<25
cardiomyopathy is part of the clinical expression of cardiac
cc/24 hr) and repeat echocardiography reveals no signif- troponin I mutations. J Clin Invest 2003; 111:209–216.
icant residual effusion. 16. Zhang J, Kumar A, Stalker HJ, et al. Clinical and molecular studies
of a large family with desmin-associated restrictive cardiomyop-
Infrequently, the tense pericardium will discharge its fluid into athy. Clin Genet 2001; 59:248–256.
the pleural space during attempts at needle passage and relieve 17. Arbustini E, Morbini P, Grasso M, et al. Restrictive cardiomyop-
tamponade. This effect is recognized on echocardiography, and athy, atrioventricular block and mild to subclinical myopathy in
patients with desmin-immunoreactive material deposits. J Am
may obviate further attempts at pericardiocentesis. While the
Coll Cardiol 1998; 31:645–653.
majority of pericardial effusions can be treated percutaneously, 18. Santamore WP, Bartlett R, Van Buren SJ, et al. Ventricular cou-
some still require subxyphoid surgical drainage, especially if pling in constrictive pericarditis. Circulation 1986; 74:597–602.
they are viscous or loculated. Bacterial infections in the peri- 19. Talreja DR, Nishimura RA, Oh JK, et al. Constrictive pericarditis
cardium also usually require large drainage. Recent hemor- in the modern era: novel criteria for diagnosis in the cardiac
rhage into the pericardium may result in clot formation that can catheterization laboratory. J Am Coll Cardiol 2008; 51:315–319.
[2/7/010/8:8:10] [124–135]
20. Lorell B, Grossman W. Profiles in constrictive pericarditis, restric- 28. Spodick DH. Acute cardiac tamponade. N Engl J Med 2003;
tive cardiomyopathy, and cardiac tamponade. In: Grossman W, 349:684–690.
ed. Cardiac Catheterization and Angiography. Philadelphia: Lea 29. Wu LA, Lapeyre AC III, Cooper LT. Current role of endomyo-
& Febiger, 1986:427–445. cardial biopsy in the management of dilated cardiomyopathy and
21. Hurrell DG, Nishimura RA, Higano ST, et al. Value of dynamic myocarditis. Mayo Clin Proc 2001; 76:1030–1038.
respiratory changes in left and right ventricular pressures for the 30. Shields RC, Tazelaar HD, Berry GJ, et al. The role of right
diagnosis of constrictive pericarditis. Circulation 1996; 93:2007–2013. ventricular endomyocardial biopsy for idiopathic giant cell myo-
22. Bush CA, Stang JM, Wooley CF, et al. Occult constrictive pericarditis. J Card Fail 2002; 8:74–78.
cardial disease. Diagnosis by rapid volume expansion and correc- 31. Tsang TS, Freeman WK, Sinak LJ, et al. Echocardiographically
tion by pericardiectomy. Circulation 1977; 56:924–930. guided pericardiocentesis: evolution and state-of-the-art tech-
23. Antman EM, Cargill V, Grossman W. Low-pressure cardiac tam- nique. Mayo Clin Proc 1998; 73:647–652.
ponade. Ann Intern Med 1979; 91:403–406. 32. Menghini VV, Savcenko V, Olson LJ, et al. Combined immuno-
24. Sagrista-Sauleda J, Angel J, Sambola A, et al. Low-pressure cardiac suppression for the treatment of idiopathic giant cell myocarditis.
tamponade: clinical and hemodynamic profile. Circulation 2006; Mayo Clin Proc 1999; 74:1221–1226.
114:945–952. 33. Drummond JB, Seward JB, Tsang TS, et al. Outpatient two-dimen-
25. Armstrong WF, Schilt BF, Helper DJ, et al. Diastolic collapse of the sional echocardiography-guided pericardiocentesis. J Am Soc
right ventricle with cardiac tamponade: an echocardiographic Echocardiogr 1998; 11:433–435.
study. Circulation 1982; 65:1491–1496. 34. Isselbacher EM, Cigarroa JE, Eagle KA. Cardiac tamponade com-
26. Burstow DJ, Oh JK, Bailey KR, et al. Cardiac tamponade: charac- plicating proximal aortic dissection. Is pericardiocentesis harmful?
teristic Doppler observations. Mayo Clin Proc 1989; 64:312–324. Circulation 1994; 90:2375–2378.
27. Shabetai R. The pathophysiology of cardiac tamponade. Cardio-
vasc Clin 1976; 7:67–89.
[2/7/010/9:43:7] [136–151]
13
Endomyocardial biopsy—indications and procedures

Gregg F. Rosner, Garrick C. Stewart, and Kenneth L. Baughman{
INTRODUCTION In the early 1970s, Caves made a series of modifications to

Disorders of the heart muscle remain among the most poorly the Konno biopsy forceps to allow percutaneous biopsy through
understood disease processes in all of cardiovascular medicine. the right internal jugular vein with only local anesthesia and
Endomyocardial biopsy techniques have now been available rapid serial tissue sampling through a preformed sheath (7,8).
for over 50 years to evaluate underlying primary myocardial Pioneered at Stanford University, the Stanford-Caves-Schulz
pathology. Cardiac biopsy has been particularly useful in bioptome became central to monitoring cardiac transplant
establishing the diagnosis and prognosis for patients with patients for rejection and served as the standard device for
recent onset or rapidly deteriorating cardiomyopathy and in endomyocardial biopsy from 1975 to 1995. The biopsy forceps
monitoring patients for rejection after cardiac transplantation. were more flexible and had features allowing the operator to
The role of endomyocardial biopsy continues to evolve as novel adjust the force applied to the forceps using a surgery-like clamp.
molecular and genetic analyses are being performed on biopsy The Stanford-Caves-Schulz device was also reusable. This led to
specimens. This chapter will review the history of endomyo- the requirement for frequent retooling and resharpening, along
cardial biopsy, define the anatomic considerations and basic with concerns about protecting patients from infection and
biopsy technique, and discuss the indications, complications pyrogen reaction. Richardson in 1974 and Kawai in 1977 added
and future directions of this important procedure. special features to the bioptome for right or left ventricular
sampling by increasing sheath flexibility, electrocardiographic
monitoring, and intracardiac maneuverability (9,10).
History of Endomyocardial Biopsy Modern bioptomes in use since the mid-1990s are dis-
Techniques for biopsying cardiac tissue outside the operating
posable, single-use devices eliminating concerns about disease
room have been available now for 50 years. In 1958, Weinberg,
transmission, pyrogen reactions, or cutting edge resharpening
Fell and Lynfield first performed heart biopsies through a small
(Fig. 13.1). Bioptomes are made in standard lengths of 50 cm for
incision in the left intercostal space. After dissection, samples of
use from the neck and chest central venous system, or over
pericardium, epicardium, and myocardium were obtained,
100 cm for use from the femoral vein or artery. The 50-cm
revealing inflammatory and restrictive cardiac disorders, and
bioptomes may be preshaped to facilitate transit across the
tubercular and traumatic causes of pericardial constriction (1).
tricuspid valve or unshaped to be inserted through a preformed
Because of the need for open incision and surgical extraction,
sheath. The preformed sheath is generally guided into the
the technique was not widely adopted. The first closed percu-
ventricular cavity over an angled pigtail or balloon flotation
taneous biopsy was performed by Sutton and Sutton in 1960
catheter. This sheath remains in the ventricular cavity through
using a needle introduced through the chest wall at the point of
the biopsy procedure, increasing risk of arrhythmia or perfo-
maximal impulse to sample myocardial tissue at the left ven-
ration and reducing operator control of the bioptome’s path
tricular apex (2). This method and other early percutaneous
and biopsy site. In contrast, preshaped bioptomes are intro-
techniques had only modest diagnostic yield and were plagued
duced through a short venous sheath and give operators
by high complication rates, including cardiac tamponade,
greater control of biopsy site selection. The degree of bioptome
arrhythmia, and postpericardiectomy syndrome (3,4).
curvature may be modified to facilitate entry into either the
In 1965, Bulloch introduced the concept of passing a biopsy
right or left ventricles. Disposable bioptomes and sheaths are
needle through the jugular vein, which allowed sampling of the
available for use from the right or left jugular, subclavian and
right interventricular septum using a cutting blade (5). Though
femoral veins as well as femoral arteries, and vary in length,
no longer in use, this technique established several principles
degree of angulation, diameter and jaw size.
central to endomyocardial biopsy today, including: use of the
right internal jugular vein, designation of the heart boundaries by
right heart catheterization prior to biopsy attempt, rotation of a ANATOMIC CONSIDERATIONS
curved biopsy sheath to avoid trauma to the tricuspid valve or Vascular Access
coronary sinus, and appropriate angulation of the biopsy device Right Internal Jugular Vein
in the right interventricular septum. Sakakibara and Konno The right internal jugular vein is the most common site for
introduced a flexible bioptome with sharpened cups allowing performance of right ventricular endomyocardial biopsy pro-
biopsy by a pinching technique rather than advancement of a cedures (11). With the patient’s head turned to the left 308 to
cutting needle (6). The original Konno bioptome is now infre- 458, the internal jugular vein is located lateral to the carotid
quently used because of its lack of durability and stiff, large artery within the anterior triangle of the neck formed by the
caliber shaft often requiring venous cut-down for access. sternal and clavicular heads of the sternocleidomastoid muscle
and the clavicle (Fig. 13.2). This triangle may be clearly outlined
{ by having the patient raise their head off the table briefly to
Deceased.
[2/7/010/9:43:7] [136–151]
ENDOMYOCARDIAL BIOPSY—INDICATIONS AND PROCEDURES 137
Figure 13.1 Modern bioptome.
Figure 13.2 Surface anatomy of the neck and jugular vein

access. Source: Courtesy of Leon Schlossberg, Ref. 12.
tense the muscular boundaries. Entry into the jugular vein in

the upper third of this triangle, well above the clavicle, will
lessen the risk of pneumothorax and allow for easier compres-
sion of accidental carotid punctures as well as the venous
access site after the procedure. Routine use of ultrasonography
is encouraged to identify the location and size of the jugular
vein prior to access attempt, particularly in patients with
challenging surface anatomy (Fig. 13.3). The jugular vein is
lateral to the carotid artery, is easily compressible when pres-
sure is applied to the ultrasound transducer, and lacks the
pulsatility of the artery, which may be confirmed by color or
pulse wave Doppler (13). Use of sonography during venous
access has been shown to improve access time and decrease
complication rates (14).
Once anatomic landmarks have been identified, the
patient’s neck is prepared with standard antibacterial solutions
and the region is isolated with sterile towels or drapes with the
patient’s head resting in a comfortable position. In patients
with low venous pressure or a small jugular vein, the legs may
be elevated or the patient placed in Trendelenberg position to Figure 13.3 Ultrasound visualization of the internal jugular vein
increase jugular venous filling and augment the puncture tar- (lateral) (right ) and the carotid artery (medial) (left ).
get. A 22- or 25-guage needle is used to introduce local
[2/7/010/9:43:7] [136–151]
anesthetic (2% lidocaine) intradermally and subcutaneously may be required. In both the internal jugular and subclavian
along the planned route of access needle entry. After successful techniques, fluoroscopy should be used to ensure that the
local anesthesia is applied, a small 1- to 2-mm superficial guidewire is directed down the vena cava toward the right
incision is made at the anticipated entry site using a surgical atrium rather that superiorly to the head.
blade and may be expanded using a mosquito clamp. This
incision prevents the venous sheath from meeting resistance Femoral Vein and Artery
when passing through the skin. Despite relatively easy cannulation, biopsy from the femoral
In the classic approach, the 22-guage needle is directed in vein can be challenging (15). The femoral vein is located medial
small increments toward the venous entry site at an angle of 308 to the femoral artery and the entry site should be below the
to 408 from vertical and 208 right of the sagittal plane. This aims inguinal ligament. The Amplatz, Seldinger, or micropuncture
the finder needle away from the more medially located carotid techniques may be used for venous access. A guide sheath is
artery. After each advancement step, the needle should be introduced into the inferior vena cava via the femoral vein. The
aspirated before infusing small amounts of lidocaine. Excess femoral artery may be accessed in a similar fashion and is the
anesthetic infiltration is discouraged as it may compress the site of entry for most left ventricular biopsy attempts. Left
jugular vein, obscure landmarks, or infiltrate into the carotid ventricular biopsies, though rare, may be indicated in patients
sheath or vocal cords resulting in transient Horner syndrome or with specific intraventricular masses or isolated left ventricular
hoarseness. Once venous blood is aspirated, the operator notes pathology, such as myocarditis or infiltrative disease (16). After
the position and an 18-guage single wall puncture needle with femoral arterial sheath insertion, a constant infusion of hepari-
syringe attached is advanced along the prior pathway of the nized saline should be maintained through the sheath to pre-
finder needle. Continuous aspiration should be applied as the vent thrombus formation within these long catheters.
18-guage needle is advanced in small increments until there is a
“give” of the vein wall and blood return is evident. A J-tipped
guidewire is introduced followed by the appropriate sheath.
Alternatively, we now frequently use a 22-guage micro-
FUNDAMENTALS
puncture needle as the probe and entry device for jugular
Biopsy Techniques
Guidance Methodology
venous access. This needle is very atraumatic and accepts a
Endomyocardial biopsies are most easily performed under
0.021-in mandril guidewire over which a coaxial 5-French
fluoroscopic guidance to define the heart borders and easily
double dilator guide sheath is advanced. Once this has entered
visualize the course of the sheath and/or bioptome into the
the jugular vein and superior vena cava, the inner cannula
heart. Some investigators advocate the use of two-dimensional
and guidewire are withdrawn and a conventional guidewire
echocardiography to guide biopsy to reduce radiation exposure
(0.038 in) is inserted through the outer cannula. This remaining
and risk of perforation (17). Echocardiographic guidance may
cannula is then removed and a 7- or 8-French self-sealing
be particularly useful in biopsying intracardiac masses in either
sheath is introduced over the guidewire. This sheath may be
the left or right heart (18). It is often technically challenging to
preformed for right ventricular biopsy, or standard length to
obtain adequate windows and visualize the biopsy forceps with
allow for preshaped bioptome use. Once the sheath is appro-
echocardiography. Compared with two-dimensional echocar-
priately positioned, the guidewire may be removed and the
diography, fluoroscopy provides the operator with superior
sheath may be aspirated, flushed and then is available for
information about the course of the bioptome and biopsy site.
endomyocardial biopsy.
Widespread use of newer technologies such as cardiac mag-
If initial attempts at venous access are unsuccessful, the
netic resonance imaging may allow the detection of a focal
probing or micropuncture needle should be retracted to just
disease process in the right or left ventricle. This information
beneath the skin and redirected more laterally. If there is still no
can then be used to direct endomyocardial biopsy to a location
venous return, medial redirection toward the carotid may be
most likely to demonstrate underlying pathology (19). Advan-
attempted. Should arterial puncture occur at any point, the
ces in three-dimensional echocardiography may improve visu-
probing needle or micropuncture needle syringe will fill with
alization of myocardium during biopsy, thereby reducing the
well-oxygenated blood. The needle must be removed immedi-
reliance of fluoroscopic imaging (20).
ately and the area compressed for five minutes or until hemo-
stasis is achieved. This problem may be avoided by using
simultaneous ultrasonographic guidance during puncture Right Internal Jugular Vein Approach: Preshaped Bioptome
attempt. The preshaped 50-cm bioptome is introduced through the
venous sheath with the bioptome tip directed toward the lateral
Right Subclavian Vein wall of the right atrium. The bioptome is slowly advanced,
The right subclavian vein may be used in those rare cases and in the mid right atrium it is rotated counterclockwise to
where anatomic abnormalities preclude use of the internal facilitate passage across the tricuspid valve, avoiding the cor-
jugular or femoral vein approaches. The entry point should onary sinus and tricuspid apparatus. The bioptome tip and
be more lateral than for routine subclavian venous access and handle have concordant motion and angulation, but position-
should be from an infraclavicular approach past the bend of the ing should always be confirmed fluoroscopically. Continued
clavicle. This more lateral approach is required so that the advancement and counterclockwise rotation allow passage into
subclavian to vena caval angle will not be too acute, preventing the mid right ventricle with the bioptome forceps directed
the relatively stiff bioptome from being positioned into the toward the septum (Fig. 13.4). Extreme care should be taken to
right heart. Application of local anesthetic and needle entry are avoid perforation of the relatively thin right atrium, vena cava,
similar to internal jugular vein access. A standard single wall or right ventricle with the stiff bioptome. If there is any resis-
needle or micropuncture kit may be used. If cannulation is tance to bioptome passage, it should be withdrawn slightly and
unsuccessful, a more inferior approach with a steeper angle a different angle of entry attempted. Bioptome forceps must
[2/7/010/9:43:7] [136–151]
Figure 13.4 Bioptome position for right ventricular septal

biopsy from the internal jugular vein approach. Abbrevia-
tions: RA, right atrium; RV, right ventricle. Source: Courtesy
of Leon Schlossberg, Ref. 12.
never be forced or prolapsed into the ventricle. If passage into withdrawn 1 to 2 cm, opened, and advanced slowly to engage
the right ventricle remains difficult, the path across the tricuspid the septum. The forceps head is then closed slowly to collect the
valve may be defined by the passage of a balloon-tipped endomyocardial specimen. Given the trebeculations within the
pulmonary artery catheter. ventricle, gentle forward pressure should be applied during
Once the bioptome is in the right ventricle, the tip should jaw closure to maintain contact with myocardium. Septal
lie against the interventricular septum. On anterior-posterior engagement may be marked by transmission of ventricular
fluoroscopy the bioptome should lie across the vertebral bodies pulsations in posttransplant patients or in those with restrictive
and is usually directed inferiorly below and to the left of the cardiomyopathy. Patients with idiopathic dilated cardiomyo-
tricuspid valve plane. Bioptome position may be confirmed by pathies may have no engagement sensation and contact with
fluoroscopy in the 308 right anterior oblique and the 608 left the septum can be confirmed only by the presence of premature
anterior oblique projection. These views will confirm that the ventricular beats.
bioptome is on the ventricular side of the atrioventricular The bioptome jaws must be closed to perform the biopsy
groove and pointed toward the septum. The absence of ectopy and pressure must be maintained on the forceps closure device
or fluoroscopy showing a position within the atrioventricular after the sample has been obtained and during withdrawal
groove suggests the bioptome has entered the coronary sinus from the right ventricle, atrium, vena cava, and sheath to
and must be withdrawn and repositioned before any biopsy is prevent the jaws from opening (Fig. 13.6). There may be a
attempted. Even within the right ventricle, the thin right slight release of traction once the specimen is removed from the
ventricular free wall must be avoided by directing the bio- septum. Excessive resistance in bioptome withdrawal suggests
ptome toward the septum (Fig. 13.5). The interventricular entrapment on the tricuspid valve apparatus or an area of scar
septum lies on a plane 458 from the chest wall, corresponding tissue. When this occurs, the forceps jaws should be opened,
to a bioptome handle orientation posteriorly and to the left. In the bioptome withdrawn, and the bioptome repositioned to
patients with right ventricular enlargement the handle may be secure another biopsy site. During routine biopsy it is not
straight posterior. uncommon for patients to experience a tugging sensation dur-
Contact with the interventricular septum is confirmed by ing specimen acquisition. Sharp chest pain, however, implies
premature ventricular beats. The biopsy forceps should be cardiac perforation. Other evidence that perforation may have
[2/7/010/9:43:7] [136–151]
Figure 13.5 Right ventricular free wall. Abbreviations: PA,

pulmonary artery; RA, right atrium; RV, right ventricle.
Source: Courtesy of Leon Schlossberg, Ref. 12.
Figure 13.6 Bioptome position and tis-

sue sample acquisition from the internal
jugular approach. Source: Courtesy of
Leon Schlossberg, Ref. 12.
[2/7/010/9:43:7] [136–151]
occurred includes persistence of premature beats, excessive exits the preformed sheath to avoid perforation. After slow
retraction of the ventricular septum during biopsy, and a sam- advancement to the septum, the jaws are slowly closed while
ple that floats in fixative, implying epicardial fat content. Any forward pressure is maintained. If the sheath tip is resting
of these clues should prompt close hemodynamic monitoring against the septum, the bioptome jaws can be exposed by
and fluoroscopy of the heart borders to detect tamponade. retracting the sheath while maintaining the forceps in a stable
Perforation is less likely in patients with prior cardiac surgery position. After the forceps are withdrawn, the sheath can be
and advanced cardiomyopathy, and more so in nonsurgical repositioned within the ventricle for another biopsy attempt.
patients with normal chamber size and systolic function. At the
end of every biopsy procedure, the heart borders should be Left Ventricular Biopsy: Femoral Artery Preformed Sheath
examined with fluoroscopy to exclude tamponade prior to The femoral artery approach usually requires insertion of
removal of venous access. 9-French self-sealing sheath through which a preformed 7-
French biopsy sheath is inserted to allow bioptome sheath
Right Internal Jugular Vein: Preformed Sheath manipulation. All femoral artery sheaths must be maintained
In this method, the preformed sheath rather than the bioptome with continuous infusion of heparinized saline to avoid
itself is advanced into the right ventricle. The sheath directs a embolic phenomenon. The preformed sheath is gently inserted
flexible straight bioptome to the desired biopsy site. A 7-French into the left ventricle across the aortic valve using a guidewire
45-cm preformed sheath may be inserted in the internal jugular and pigtail catheter. Once in the left ventricle, areas of prior
vein. The performed sheath is positioned into the right ventricle myocardial infarction and inferior-posterior positions should
with the aid of a guidewire or balloon-tipped catheter. Once in be avoided to reduce the perforation risk in these relatively
the right ventricle, the guidewire or catheter is removed and the thin-walled sites. After the sheath is cleared of debris by aspi-
sheath flushed with heparinized saline and connected to an ration and flushing, a 104-cm bioptome is inserted into the left
infusion port to maintain patency. The sheath should be free- ventricular cavity. The biopsy forceps should be directed away
floating and not abutting ventricular myocardium. The distal from the mitral valve apparatus. The jaws are opened upon
segment of a flexible bioptome should be angulated or curved exiting the sheath, directed toward the ventricular wall, the
before insertion to avoid straightening the preformed sheath. specimen is encapsulated, and the jaws are closed firmly while
Precurved bioptomes may also be used through the preshaped extracting the sample. The sheath is maintained within the left
sheath. The bioptome jaws should be opened immediately ventricle and repositioned to ensure sampling from several
upon exiting the sheath to reduce risk of perforation. The different sites. Since most myocardial disease processes affect
bioptome is directed posteriorly, perpendicular to the interven- both ventricles, right ventricular biopsy is preferred because of
tricular septum. Gentle pressure is applied as the jaws are greater ease, short procedure time, and reduced likelihood of
closed, then the sample may be removed as described above. morbidity. In diseases with confirmed selective left ventricular
The sheath should be flushed with saline and repeat biopsy involvement or in patients in whom right ventricular biopsy
performed as indicated. has been unsuccessful, left ventricular biopsy may be
attempted. In general, left ventricular biopsy is reserved for
Femoral Vein Approach: Preformed Sheath cases with selective left ventricular involvement such as in
From the femoral vein, a 7-French preformed guiding sheath endomyocardial fibrosis, scleroderma, left heart radiation,
can be introduced into the right ventricle using a balloon- and cardiac fibroelastosis of infants and newborns.
tipped catheter, pigtail catheter, or guidewire. Rarely, in chil-
dren, the femoral venous approach may be used to facilitate left Postprocedure Care
ventricular biopsy by transseptal puncture (21). The standard After biopsy sheath removal, pressure must be applied to
preformed femoral sheath has a 1308 angle from the right atrium prevent local bleeding complications. After uncomplicated
into the right ventricle to facilitate positioning. The preshaped biopsy, patients with biopsies from the jugular vein may be
sheath may increase the risk of perforation because the operator discharged home after one hour of observation. Patients who
may have less control of the direction of the bioptome to the had femoral venous entry require two to three hours of supine
cardiac septum. The femoral sheath must be evaluated before bed rest before safely attempting ambulation. Patients with
insertion to ensure that the sheath length after the angulation femoral arterial access require several hours of bed rest with
does not exceed the anatomic distance from the right atrium to or without use of a vascular closure device before ambulation.
ventricle for a given patient because of risk of perforation. This All patients must be monitored for bleeding and hemodynamic
may be confirmed by placing the sheath on the exterior of the changes. The bandage applied to the vascular access site may
patient on a sterile field and imaging under fluoroscopy. Once be removed in 24 hours and oral intake may resume once
the sheath is inserted it should be flushed to avoid clot forma- patients can sit up.
tion. If there is any question about the position of the sheath tip,
hand injection of contrast dye and hemodynamic monitoring
may be helpful. After sheath insertion, the unshaped bioptome INDICATIONS
should be curved, similar to the jugular approach, to avoid Transplant Rejection
straightening the preformed sheath when inserted. Posterior Endomyocardial biopsy has been the cornerstone of monitoring
angulation of the bioptome tip out of the plane relative to the patients for rejection after heart or heart-lung transplantion
broad proximal curvature of the sheath can help direct the (22,23). Early rejection may be detected on endomyocardial
bioptome toward the interventricular septum upon exiting biopsy before clinical manifestations. Regular posttransplantation
the sheath (Fig. 13.7). biopsies monitor antirejection therapy and can confirm the ade-
Fluoroscopy can be used to track bioptome passage quacy of pulsed immunosuppressive therapy for episodes of
through the sheath and can confirm interventricular septal acute rejection. Despite promising research, no single methodol-
position. Bioptome jaws should be opened just as the bioptome ogy has had the predictive accuracy to replace endomyocardial
[2/7/010/9:43:7] [136–151]
Figure 13.7 Endomyocardial biopsy from the

femoral vein. Abbreviation: RV, right ventricle.
Source: Courtesy of Leon Schlossberg, Ref. 12.
biopsy in the assessment of cellular and humoral rejection sis, which is known only after biopsy. Few randomized, con-
(24–26). Because rejection is a diffuse immune-mediated phenom- trolled treatment trials have evaluated the utility of heart
enon, sampling errors are rare. The severity of rejection on biopsy biopsy in disease management, and recommendations are
has been divided into four grades as established by the Interna- made on the basis of case-control series and expert opinion.
tional Society of Heart and Lung Transplantation (27). Grade 0R A recent comprehensive Scientific Statement from the Ameri-
represents no rejection, grade 1R mild rejection, and grade 2R can Heart Association, American College of Cardiology, and
moderate rejection. Severe rejection (grade 3R) is marked by European Society of Cardiology outlines the appropriate role of
multifocal aggressive infiltrates and/or myocyte damage or dif- endomyocardial biopsy outside the posttransplant setting (28)
fuse polymorphous infiltrate with necrosis and a variable degree (Table 13.1). Biopsy does have an important role in the evalu-
of edema, hemorrhage, or vasculitis. More severe rejection requires ation of unexplained heart failure (29–31). Heart failure is
aggressive immunosuppression even in asymptomatic patients. considered to be unexplained when comprehensive testing
including ECG, chest radiography, echocardiography, and cor-
onary angiography fails to reveal a diagnosis. Standard evalu-
Biopsy in the Management of Cardiovascular ations of patients with new-onset cardiomyopathy are
Disease informative, but as many as half of this population has no
Routine use of endomyocardial biopsy to inform management diagnosis after routine testing (32). The potential value of direct
of a variety of cardiovascular diseases in patients without a assessment of heart muscle tissue in patients with new-onset
heart transplant remains controversial. Decisions to proceed heart failure may be more valuable in this population.
with endomyocardial biopsy are most often made on the basis Patients most appropriate for biopsy are those with
of clinical presentation, not the underlying pathologic diagno- new-onset heart failure of less than two weeks duration
[2/7/010/9:43:7] [136–151]
Table 13.1 Recommendations for Endomyocardial Biopsy

Class I: Conditions for which there is evidence or general agreement that biopsy is beneficial, useful and effective
l New-onset heart failure of <2 wk duration associated with a normal-sized or dilated left ventricle and hemodynamic compromise.
l New-onset heart failure of 2 wk to 3 mo duration associated with a dilated left ventricle and new ventricular arrhythmias, second- or third-
degree heart block, or failure to respond to usual care within 1–2 wk
Class IIa: Conditions for which the weight of evidence/opinion is in favor of usefulness/efficacy
l Heart failure of >3 mo duration associated with a dilated left ventricle and new ventricular arrhythmias, second- or third-degree heart
block, or failure to respond to usual care within 1–2 wk
l Heart failure associated with a dilated cardiomyopathy of any duration association with suspected allergic reaction and/or eosinophilia
l Heart failure associate with suspected anthracycline cardiomyopathy
l Heart failure associated with unexplained restrictive cardiomyopathy
l Suspected cardiac tumors
l Unexplained cardiomyopathy in children
Class IIb: Conditions for which usefulness/efficacy of biopsy is less well established by evidence/opinion
l New-onset heart failure of 2 wk to 3 mo duration associated with a dilated left ventricle, without new ventricular arrhythmias, second- or
third-degree heart block, that responds to usual care within 1–2 wk
l Heart failure of >3 mo duration associated with a dilated left ventricle and new ventricular arrhythmias, second- or third-degree heart
block, that responds to usual care within 1–2 wk
l Heart failure associated with unexplained hypertrophic cardiomyopathy
l Suspected arrhythmogenic right ventricular dysplasia/cardiomyopathy
l Unexplained ventricular arrhythmias
Class III: Conditions for which there is evidence and/or general agreement that biopsy is not useful
l Unexplained atrial fibrillation
accompanied by hemodynamic compromise or those with heart Table 13.2 Cardiac Disorders with Specific Findings on Biopsy
failure of two weeks to three months duration who have heart
Immune or inflammatory disease states
failure or dysrhythmias that fail to respond to standard thera- Myocarditis
pies. Patients with severe cardiac compromise and duration of Cardiac allograft rejection
less than two weeks often have fulminant lymphocytic myo- Sarcoidosis
carditis and a good prognosis (33–35). These cases must be Cytomegalovirus infection
distinguished from similarly aggressive disorders such as giant Toxoplasmosis
cell myocarditis and necrotizing eosinophilic myocarditis, Rheumatic carditis
which have a similarly fulminant course, often requiring per- Chagas disease
manent mechanical support or consideration of cardiac trans- Kawasaki disease
plantation (36,37). Giant cell myocarditis in particular may be Degenerative
Idiopathic cardiomyopathy
associated with both ventricular tachycardia and atrioventric-
Anthracycline cardiomyopathy
ular block (38). Confirmation of histologic diagnosis of either Radiation cardiomyopathy
giant cell or necrotizing eosinophilic myocarditis would lead to Infiltrative
immunosuppressive treatment, while fulminant lymphocytic Amyloidosis
myocarditis resolves without such agents. Studies are ongoing Gaucher’s disease
to define the appropriate diagnostic criteria for myocarditis and Hemochromatosis
guide appropriate immune modulating therapies (39,40). Fabry’s disease
Patients with long-term and established dilated cardio- Glycogen storage disease
myopathy responding to usual heart failure treatments have Ischemic
less to gain from endomyocardial biopsy. In these subjects, Acute myocardial infarction
Chronic ischemic cardiomyopathy
biopsy generally displays nonspecific findings of myocyte
Schonlein-Henoch purpura
hypertrophy, interstitial and replacement fibrosis, and endo- Cancer
cardial thickening (41,42). Such findings do not establish etiology, Primary cardiac cancer
long-term prognosis, or guide specific therapies. Nevertheless, Metastatic cardiac cancer
there is a subset of patients with dilated cardiomyopathy who
may have specific disease processes identified on biopsy, partic-
ularly those who fail to respond to standard heart failure
therapies and have refractory symptoms (Table 13.2). Patients sis in patients with underlying restrictive physiology, a com-
with suspected cardiac sarcoidosis or myocarditis should be mon entity seen in patients with heart failure and a preserved
considered for heart biopsy. ejection fraction. By helping distinguish between restrictive
Endomyocardial biopsy is also recommended for patients cardiomyopathy and constrictive pericarditis, biopsy can help
with cardiomyopathy from suspected anthracycline toxicity, spare patients inappropriate medical or surgical therapies (43).
cardiomyopathy associated with allergic or eosinophilic reac- Disorders causing a restrictive cardiomyopathy include pri-
tion, patients with unexplained ventricular tachycardia or con- mary amyloidosis, Loeffler endomyocardial fibrosis, carcinoid-
duction disease, or children with unexplained cardiomyopathy related heart disease, Fabry disease, and glycogen storage
(28). Endomyocardial biopsy may also help establish a diagno- diseases (Table 13.3) (44,45).
[2/7/010/9:43:7] [136–151]
Table 13.3 Restrictive Cardiomyopathies Delayed complications include access site bleeding and dam-
age to the tricuspid valve from repeated trauma associated with
Myocardial Endocardial
transplant rejection surveillance.
Noninfiltrative Fibrosis Factors associated with endomyocardial biopsy compli-
Idiopathic Hypereosinophilic cation risk include operator experience, patient clinical status,
Familial Carcinoid indication for biopsy, access site, underlying cardiac conduc-
Hypertrophic Metastatic cancer tion disease (presence of a left bundle branch block), and
Scleroderma Radiation
possibly bioptome, since allergic reactions to reusable bio-
Diabetic Anthracyclines
Infiltrative Drugs ptomes have been reported. Fowles and Mason reported a
Amyloid complication rate of <1% in over 4000 transplant and cardio-
Sarcoid myopathy patients biopsied at Stanford University (46). They
Gaucher’s reported no deaths in their series. Cardiac tamponade occurred
Hurler’s in four patients (<0.14%), none of whom needed thoracotomy.
Fatty Other complications included atrial fibrillation in three patients,
Storage and sustained ventricular arrhythmia in one patient. Complica-
Hemochromatosis tions related to the internal jugular vein approach included
Fabry’s pneumothorax in three patients, uncomplicated air embolism in
Glycogen
six patients, transient right recurrent laryngeal nerve paresis in
two patients, and Horner’s syndrome in one patient. Similarly,
CLINICAL ASPECTS Sekiguchi and Take reported, by questionnaire, an overall com-
Equipment needed for endomyocardial biopsy is listed in plication rate of 1.17% in 6739 worldwide patients, including
Table 13.4. perforation in 28 patients (0.42%) and death in two patients
(0.03%) (47).
Deckers et al. prospectively recorded complications of
Biopsy Complications right heart biopsy from 546 procedures in 464 consecutive
Complications associated with endomyocardial biopsy are con- patients with new-onset idiopathic cardiomyopathy (48). The
sidered either acute (while the patient is still in the catheter- internal jugular vein was the primary site of access in 96% of
ization laboratory) and delayed. The vast majority of cases. An overall complication rate of 6% occurred in this
complications fall into the acute category. Potential acute series. Of these complications, 15 (2.7%) occurred during cath-
complications include ventricular perforation and pericardial eter insertion including 12 arterial punctures (2%), 2 vasovagal
tamponade, malignant ventricular or supraventricular arrhyth- reactions (0.4%), and 1 episode of prolonged bleeding (0.2%), all
mias, transient complete heart block, pneumothorax, central without sequelae. There were 18 (3.3%) complications during the
artery puncture, nerve paresis, thromboembolism, systemic actual biopsy, including six arrhythmias (1.1%), five conduction
platelet embolization associated with left ventricular biopsy, abnormalities (1%), four possible perforations (0.7%), and three
venous or arterial hematoma, damage of the tricuspid or mitral definite perforations (0.5%) (pericardial fluid). Two (0.4%) of the
valve, and creation of arterial venous fistula within the heart. three patients with definite perforation died.
Biopsy complications occur at a similar rate when echo-
Table 13.4 Equipment cardiography rather than fluoroscopy is used to guide the
procedure. Han et al. prospectively recorded complications of
18-gauge Amplatz needle or 22-gauge micropuncture needle right heart biopsy in 90 consecutive nontransplant patients who
250 mL of flush solution (with heparin unless allergic) underwent two-dimensional echocardiography-guided trans-
4- or 5-French micropuncture sheath femoral biopsy (49). The overall complication rate was 5.6%
7-, 8-, or 9-French self-sealing introducer with 0.038 guidewire
and no deaths occurred in this cohort. Myocardial perforation
Assorted syringes (1, 5, 10, 20 mL)
Automatic intermittent cutaneous or invasive blood pressure monitor occurred in three patients but did not progress to cardiac
Continuous electrocardiographic monitor tamponade requiring pericardiocentesis in any patient. Unsta-
Continuous oxygen saturation monitor ble ventricular tachycardia occurred in one patient, and a new
Defibrillator and persistent right bundle branch block occurred in one
Dry ice patient.
Emergency equipment Baraldi-Junkins et al. reviewed 2454 endomyocardial
Ether screen or drape support (optional) biopsies performed in 133 cardiac transplant patients (50). An
Formalin overall complication rate of 3.0% occurred, however, no perfo-
Glutaraldehyde rations or deaths were observed in this cohort. Of these, 56
Lidocaine 1% or 2%, 15 mL
(2.3%) complications were associated with catheter insertion,
Micropuncture wire, 0.021
Mosquito clamp or small-tipped instrument including carotid puncture (1.8%), vasovagal reaction (0.1%),
Number 11 surgical blade and handle and prolonged bleeding (0.4%). Complications during biopsy
1 25-, 1 22-, and 3–4 18-gauge needles included arrhythmias (0.25%) and conduction abnormalities
Pacemaker and pacing leads (0.2%). Additional complications included five episodes (three
Pericardiocentesis set patients) of allergic reaction to a reusable bioptome and one
Plastic or cloth drape set case of pacemaker dislodgement.
Povidone-iodone, alcohol, or both Recently, Holzmann and colleagues reported results of a
Resuscitation drugs and equipment retrospective and prospective study examining the incidence of
Tissue preservative major and minor complications of right ventricular endomyo-
Vascular ultrasound machine (i.e., SonoSite M-turbo)
cardial biopsies via the right femoral vein approach at a single
[2/7/010/9:43:7] [136–151]
center in Germany (51). In their study, 1919 patients underwent endomyocardial biopsy for the purpose of heart transplant
2505 endomyocardial biopsy procedures retrospectively eval- rejection surveillance.
uated between January 1995 and December 2003, and 496
patients underwent 543 endomyocardial biopsy procedures Perforation
prospectively assessed between January 2004 and December The greatest risk to patients undergoing endomyocardial
2005. Major complications (pericardial tamponade requiring biopsy is ventricular perforation. If hemodynamically signifi-
pericardiocentesis, pneumothorax or hemothorax, permanent cant and left uncorrected, such perforation can lead to peri-
pacemaker support, cases requiring emergency cardiac sur- cardial tamponade and death (53). Factors associated with an
gery, or death) were extremely rare in both the retrospective increased likelihood of perforation include bleeding diathesis,
study (0.12%) and the prospective study (0%). No deaths were recent receipt of heparin, pulmonary hypertension, and
reported during either study (total of 3048 endomyocardial increased right ventricular systolic pressures or right ventricu-
biopsies). Major complications reported during the retrospec- lar enlargement. Patients with a prothrombin time >18 seconds
tive study included: cardiac perforation in two cases (0.08%) or who have received heparin without reversal within the prior
and permanent pacemaker in one patient (0.04%). No major two hours should probably not undergo endomyocardial
complications occurred during the prospective study. A differ- biopsy. Perforation is usually a complication of injury to the
ence was reported in the incidence of minor complications right ventricular free wall, which is only 1 to 2 mm thick.
(hemodynamically insignificant pericardial effusion, conduc- Interestingly, performance of left ventricular biopsy shares
tion abnormalities not requiring permanent pacemaker, or similar perforation complication rates despite significantly
arrhythmias not requiring ACLS) between the retrospective thicker ventricular walls.
and prospective studies. Minor complications occurred in It is critical that clinicians who perform endomyocardial
0.20% of the endomyocardial procedures in the retrospective biopsy have a high index of suspicion for cardiac perforation.
study and 5.5% in the prospective study. Five patients (0.20%) To this end, any patient complaining of pain during the per-
developed complete heart block requiring temporary pacing formance of the endomyocardial biopsy should be considered
during the retrospective study. Minor complications seen dur- to have experienced cardiac perforation. Typically, patients
ing the prospective study included four cases (0.74%) of hemo- with perforation immediately complain of a visceral pain and
dynamically insignificant pericardial effusion, conduction within one to two minutes may develop vagal symptoms
abnormality not requiring temporary pacing in 12 patients including bradycardia and hypotension. Despite the excess
(1.84%), temporary pacing requirement in eight cases (1.47%), parasympathetic tone thought to underlie these symptoms,
and atrial fibrillation in 6 patients (1.1%). The authors believed limited benefit is achieved by atropine administration. Further
the most likely reason for the increased minor complication rate biopsy attempts are contra-indicated until a thorough investi-
observed during the prospective study compared with the gation into the patient’s complaints has been completed. If
retrospective study was a more detailed documentation sheet cardiac perforation is suspected, continuous evaluation of
initiated during the prospective study. right atrial pressure and the pulsatility of the right and left
The risk of endomyocardial biopsy in children was stud- heart borders by fluoroscopy should be performed. Increased
ied by Pophal et al. in a retrospective review of 1000 consecu- right atrial pressure and loss of pulsation of heart borders are
tive heart biopsies in 194 children (52). The mean age at the strong indicators for pericardial tamponade. Emergent trans-
time of biopsy was 8.6 years (8 days–18 years), mean weight thoracic echocardiography should be obtained to determine the
was 30 kg (2.8–127 kg), mean height was 121 cm (48–187 cm) presence and severity of pericardial blood accumulation.
and mean body surface area was 0.98 m2 (0.18–2.05 m2). Cardiovascular collapse or electrical mechanical disasso-
Indications for heart biopsy included heart transplant rejection ciation (PEA arrest) in the setting of a biopsy should be con-
surveillance (84.6%) and the evaluation of cardiomyopathy or sidered to be presumptive evidence of pericardial tamponade,
arrhythmia for possible myocarditis (15.4%). The overall inci- and mandates immediate pericardiocentesis, even in the
dence of a serious complication from endomyocardial biopsy absence of echocardiographic confirmation of tamponade.
was 1.9%. There were nine perforations (0.9%) and one death Occasionally acute bleeding into the pericardial space will
(0.1%) secondary to perforation. In the evaluation of cardiomy- clot and prevent adequate draining by pericardiocentesis.
opathy/myocarditis, the incidence of complication was 9.1%, Should this situation arise in a hemodynamically unstable
perforation was 5.2% and mortality was 0.6%. In patients patient, it may be necessary that the pericardial space be surgi-
undergoing biopsy for transplant rejection surveillance, the cally evacuated, occasionally in the catheterization laboratory.
incidence of complication was 0.6%, perforation was 0.1%, Because of the risk of tamponade, a pericardiocentesis tray
and no deaths occurred. should always be available in the procedure room where endo-
No reported series has estimated complication rates from myocardial biopsies are performed.
left ventricular biopsies. In addition to the risk of arterial versus
venous access, these patients require antiplatelet therapy and Malignant Ventricular Arrhythmias
heparinized sheaths, which may increase the risk of bleeding. Ventricular ectopy is an expected consequence of cannulation
Platelet emboli into the systemic arterial bed place patients at and mechanical stimulation of the cardiac chambers by the
increased risk for central nervous system complications. The sheath or bioptome. In fact, premature ventricular contractions
risk of perforation from left ventricular biopsy is probably not are utilized as an indication of appropriate placement of the
decreased compared with the right ventricular approach. bioptome or biopsy sheath within the ventricular cavity. Rarely,
On the basis of these studies the estimated risk of com- sustained malignant ventricular arrhythmias may develop dur-
plication related to endomyocardial biopsy for the evaluation of ing the biopsy procedure. Risk factors for this complication
cardiomyopathy or for possible myocarditis is 1% to 6%. The include cardiomyopathy and preexistent ventricular arrhyth-
risk of fatal complication is 0% to 0.4%. Of note, there appears mias. Treatment begins with immediate withdrawal of the
to be a lower risk of morbidity and mortality related to bioptome or biopsy sheath from the ventricular cavity. Should
[2/7/010/9:43:7] [136–151]
this fail to stop the arrhythmia, medical therapy with antiar- output (i.e., cardiomyopathy), this distinction can sometimes be
rhythmic agents or cardioversion may be necessary. difficult. To help distinguish between arterial and venous
blood, the operator may send a sample of blood for blood gas
Supraventricular Arrhythmias analysis. Alternatively, the operator can insert a small 18-gauge
During instrumentation of the right atrium, the atrial wall may catheter over the guidewire and then determine the pressure
be stimulated leading to supraventricular arrhythmias. Risk waveform of the vessel cannulated. Puncture of an artery
factors for this complication include prior history of supra- utilizing the finder or micropuncture needle should be
ventricular arrhythmia or elevated right sided filling pressures. addressed by withdrawal of the needle and compression of
Operators should try to avoid right atrial wall contact, partic- the vessel until homeostasis is obtained. This does not preclude
ularly in patients identified at risk. In the event that a supra- performance of a safe venous approach. Cannulation of an
ventricular tachycardia develops, mechanical interruption of artery with a large (7–9 French) sheath is a more serious
the circus rhythm may be attempted by touching the right atrial complication that requires urgent vascular surgery consulta-
wall with the bioptome. However, this may lead to an increased tion. In this situation the sheath should not be removed before
risk of cardiac perforation. surgical consultation because of the risk of hemorrhage.
Heart Block Thromboembolic Phenomenon

Occasionally patients with preexistent left bundle branch block Patients with preformed sheaths which are not continuously
may develop complete heart block during manipulation of flushed may develop a clot within the sheath during the per-
instruments within the right heart. Pressure against the septum formance of the endomyocardial biopsy (58). As the bioptome
near the tricuspid apparatus may “stun” the right bundle is advanced through the clot-containing sheath, this can lead to
resulting in a new right bundle branch block. In patients with expulsion of the clot into the patient’s circulation. When this
a preexistent left bundle branch block, the addition of a new occurs during right ventricular biopsy it can lead to pulmonary
right bundle branch block results in progression to complete embolization. Should this occur during a left ventricular biopsy
heart block. Removal of the offending bioptome or catheter or during a right ventricular biopsy in a patient with a right-to-
often resolves the complete heart block; however, should com- left shunt, systemic embolization can occur. A risk unique to
plete heart block persist, a temporary pacing catheter can be left ventricular biopsy is systemic platelet embolization. The
inserted into the right ventricular cavity. For this reason, a incidence of platelet emboli during left heart biopsy has not
temporary pacemaker and pacing wire should be immediately been reported. Some operators suggest use of antiplatelet ther-
available in the catheterization laboratory for emergent use if apy such as aspirin prior to biopsying the left heart.
needed, particularly in patients with a preexistent left bundle
branch block. Nerve Paresis
Infiltration of local anesthesia into the jugular venous or carotid
Pneumothorax and Hemothorax sheath may result in a Horner’s syndrome, vocal cord paresis,
Puncture of the lung pleura during performance of internal or rarely diaphragmatic weakness (59,60). These complications
jugular or subclavian venous access may result in a pneumo- typically last a few hours (on the basis of the half-life of the
thorax or hemothorax. On the basis of a large meta-analysis of local anesthetic used) unless direct nerve trauma has occurred
17 prospective comparative trials including data on 2085 jugu- from the needle itself.
lar and 2428 subclavian catheters, the risk of one of these
complications has been estimated at 1.5% for subclavian venous Tricuspid Valve Damage
access and 1.3% for internal jugular venous access (54). Several Tricuspid regurgitation is a well recognized complication of
strategies can be utilized to minimize this risk. A growing body endomyocardial biopsy (61). In fact, tricuspid regurgitation is
of literature supports the use of real time two-dimensional the most common valvular lesion after orthotopic heart trans-
ultrasound guidance for internal jugular venous access plantation. The most common etiology of significant tricuspid
(55,56). Secondly, strict attention to detail should be maintained regurgitation in the posttransplant patient is endomyocardial
during insertion. This includes performing a “higher” internal biopsy performed to detect allograft rejection. Endomyocardial
jugular stick with avoidance of the immediate supraclavicular biopsy can lead to direct anatomic disruption of the valve appa-
region, continuous aspiration of the needle plunger during ratus, such as ruptured chordae tendinae or torn leaflet (62). In
every attempt at venous entry, and also during subclavian their study of 98 posttransplant patients, Mielniczuk et al. found
venous access the operator should never let the needle drop histologic evidence of chordal tissue in 9% of endomyocardial
below the horizontal plane. Immediate investigation with flu- biopsy specimens, which accounted for 47% of patients with
oroscopy of the lung margins should be performed in any significant tricuspid regurgitation in their cohort (63). Nguyen
patient undergoing endomyocardial biopsy who complains of et al. examined whether there was a correlation between the
spontaneous shortness of breath. Urgent pneumothorax or number of endomyocardial biopsies and the risk of severe tri-
hemothorax evacuation should be performed as needed. cuspid valvular regurgitation (64). In their study of 101 posttrans-
plant patients they found 60% of patients with more than
Puncture of the Carotid, Subclavian, or Femoral Artery 31 endomyocardial biopsies had developed severe tricuspid
Central veins lie adjacent to their corresponding arteries. The regurgitation, whereas none of the patients with fewer than
risk of carotid, subclavian, and femoral artery puncture has 18 endomyocardial biopsies had severe tricuspid regurgitation.
been estimated to be approximately 3%, 0.5% to 5%, and 9%,
respectively (54,57). Under most circumstances, the operator Hematoma
can easily distinguish an arterial puncture from a venous A venous or arterial hematoma may form as a result of exces-
puncture by the red color and pulsatile flow of arterial blood. sive movement of the sheath during the biopsy procedure,
In patients with significant hypoxemia and/or reduced cardiac inadequate compression of the vascular access site after
[2/7/010/9:43:7] [136–151]
removal of the sheath, or late bleeding due to a transient or expert pathology panel during review of the same biopsy
sustained increase in right atrial or mean arterial pressure. samples at a later date. In a separate analysis of interobserver
Patients with coagulopathy or who are on anticoagulant ther- variability, Shane et al. submitted endomyocardial biopsy
apy as well as aspirin are at increased risk for hematoma specimens from 16 patients with dilated cardiomyopathy to
formation and should be monitored more closely. seven expert cardiac pathologists. Their assessments varied
remarkably with respect to significant fibrosis (25–69%), hyper-
Arterial-Ventricular Fistula trophy (19–88%), nuclear changes (31–94%), lymphocyte count
Occasionally arterial-ventricular fistulas develop between small per high-power field (0–38%), and the diagnosis of myocarditis.
branches of a coronary artery and the right ventricle in post Definite or probable myocarditis was diagnosed in 11 of
cardiac transplant patients (65). These are caused by inadver- 16 patients by at least one pathologist. However, of the
tent biopsy of septal branches of a coronary artery and subse- 11 patients, three of seven pathologists agreed on the diagnosis
quent arterial communication into the right ventricle. Several in 3 patients and two of seven pathologists agreed on the
studies have shown that these fistulae are of no clinical conse- diagnosis in 5 patients.
quence and can be followed conservatively. Several researchers have demonstrated the presence of
cardiotropic virus in myocardium in the absence of Dallas
criteria myocarditis. Martin et al. utilized polymerase chain
LIMITATIONS reaction (PCR) to analyze myocardial tissue samples from
Sampling Error 34 patients with suspected acute viral myocarditis. They dem-
One limitation of endomyocardial biopsy as a diagnostic tool onstrated that 26 heart biopsy samples were positive for viral
for myocarditis or transplant rejection is that it is prone to pathogens, and 13 of the 26 positive samples had no evidence
sampling error. Since myocarditis, and to a lesser extent trans- of Dallas criteria myocarditis (70). Pauschinger et al. found
plant rejection, tend to be focal processes, accurate diagnosis either adenoviral or enteroviral PCR positivity in 24 myocardial
depends on adequate sampling of the myocardium. Standard tissue samples (none of which showed histopathologic evi-
bioptomes sample approximately 1 to 2 mm3 (30 mg) of dence of myocarditis) from 94 patients with “idiopathic”
endomyocardium with each biopsy. Researchers have demon- dilated cardiomyopathy (71). In a separate study of 45 patients
strated on ex-vivo hearts with histologically proven myocardi- with left ventricular dysfunction and suspected myocarditis,
tis (either postmortem or explanted) that sampling error Paushchinger et al., demonstrated nonreplicative enterovirus in
contributes appreciably to false negative results (66). Chow 40% of patients. Of the 18 patients with nonreplicative virus,
et al. demonstrated that from a single endomyocardial biopsy, 56% were found to have active viral replication as well (72). In
histologic myocarditis could be demonstrated in only 25% of this study 13% of the biopsy samples were diagnosed as having
samples (67). With more than five random samples, Dallas Dallas criteria borderline myocarditis, however, histopathology
criteria myocarditis could be diagnosed in approximately did not help to distinguish between patients with and those
two-thirds of subjects. Most recently, Mahrholdt et al. demon- without enteroviral positivity. Why et al. showed in their
strated by cardiovascular magnetic resonance imaging that the cohort of 120 patients with idiopathic dilated cardiomyopathy
earliest myocardial inflammatory abnormalities in myocarditis that the 34% who were enteroviral positive had a significantly
are located in the lateral wall of the left ventricle, a site that is worse outcome over two years compared with those who were
not available to biopsy with the standard approach (19). For enteroviral negative (73). Taken together, virus can exist in the
transplant rejection surveillance, the sensitivity of detecting myocardium (even in a replicative form) in the absence of
transplant rejection approaches 98% when five adequate biopsy histopathologic findings adequate to meet Dallas criteria and
samples are obtained (68). may adversely affect outcome.
Another limitation in the usefulness of the Dallas criteria
is the dissociation between histopathology findings and
Limitations of the Dallas Criteria response to immune modulation therapy. For instance, in the
Originally proposed in 1986, the Dallas criteria established a Myocarditis Treatment Trial, there was no difference in one- or
histopathologic categorization by which the diagnosis of myo- five-year survival or improvement of left ventricular ejection
carditis could be made (69). According to the Dallas criteria, fraction at 28 weeks in patients with Dallas criteria myocarditis
active myocarditis requires an inflammatory infiltrate and treated with immunosuppressive therapy or placebo. Other
associated myocyte necrosis or damage uncharacteristic of an authors have used alternative criteria to diagnose immune-related
ischemic event. Borderline myocarditis requires a less intense heart disease such as HLA upregulation on endomyocardial
inflammatory infiltrate and no light microscopic evidence of biopsy. Wojnicz et al. found HLA upregulation in cardiac biopsy
myocyte destruction. Data from the Myocarditis Treatment samples from 84 of 202 patients (41.6%) with new-onset cardio-
Trial reveal that approximately 10% of patients (214 out of myopathy, while only 27% were positive by Dallas criteria for
2233) with clinically suspected myocarditis (new-onset unex- myocarditis (74). HLA-identified patients were randomized to
plained heart failure during the two years preceding enroll- receive treatment with either immunosuppressive therapy or
ment) who underwent endomyocardial biopsy were diagnosed placebo. After two years of follow-up, there was no significant
by the current histopathologic Dallas criteria (40). difference in the primary endpoint (a composite of death, heart
To compound this further, the Dallas criteria require that transplantation, and hospital readmission) between the study
biopsy specimens be examined by qualified cardiac patholo- groups (22.8% for the immunosuppression group and 20.5% for
gists. Additionally, even when specimens are examined by the placebo group); however, the ejection fraction in the immu-
expert cardiac pathologists there are variations in the interpre- nosuppressive group increased from 24% to 36%, whereas it
tation of histologic samples. In the Myocarditis Treatment Trial, remained unchanged in the placebo group (25–27%).
only 64% of the 111 patients diagnosed with myocarditis by Despite the presence of Dallas criteria myocarditis,
endomyocardial biopsy had their diagnosis confirmed by the response to treatment may be influenced by the presence of
[2/7/010/9:43:7] [136–151]
virus or immunologic response to infection. This was most RNAlater solution and stored at 808F. These study samples
clearly demonstrated by Frustaci et al. in their study of immu- may be required for suspected myocarditis, amyloid classifica-
nosuppressive therapy in patients with Dallas criteria myocar- tion, tumor typing, or viral genome analysis. Additional sample
ditis who failed to respond to conventional therapy (75). Out of preparation may be individualized for evaluation of specific
652 patients who underwent endomyocardial biopsy, 112 were disease states (e.g., amyloid, iron staining). It is the operator’s
identified with Dallas criteria myocarditis. Of these 112 responsibility to ensure timely delivery of biopsy samples to
patients, 41 had progressive congestive heart failure despite the appropriate pathologic laboratory.
standard medical therapy and were treated with prednisone
and azathioprine for six months. Twenty-one patients
responded to treatment with improvement of left ventricular
Cardiac Pathologist
Experienced cardiac pathologists are central to any biopsy
ejection fraction from 25.7% to 47.1%, and showed evidence of
program. The safest and most pristine biopsy specimen is
healed myocarditis on follow-up biopsy. Twenty patients failed
useless without an experienced cardiac pathologist who is
to respond to treatment and showed a histologic evolution
fully trained in the evaluation of biopsy-obtained tissue and
toward dilated cardiomyopathy. Viral genomes were present
conversant with the latest classification schemes. Crush arti-
in endomyocardial biopsy specimens in 85% of nonresponders
facts or contraction bands are frequently present in biopsy
versus 15% of responders. Circulating cardiac antibodies were
specimens and may be incorrectly interpreted by an inexper-
present in 90% of responders versus 0% of nonresponders.
ienced or noncardiac pathologist. The operator may assist the
Given that the Dallas criteria are prone to sampling error,
pathologist through careful handling of the biopsy specimen in
interobserver variability, variance with other markers of viral
the catheterization laboratory, by ensuring that the heart biopsy
infection and immune activation in the heart, and variance with
specimens obtained are delivered to the appropriate laboratory
treatment outcomes, we suggest that the Dallas criteria should
for analysis, and by reviewing the slide material obtained with
no longer be used to diagnose myocarditis in isolation. Instead,
the pathologist to provide clinical details and to ensure that
myocarditis should be diagnosed on the basis of a combination of
special studies are obtained as needed.
clinical presentation, histopathology, immunohistochemistry,
viral PCR, cardiac antibody assessment, and imaging results.
Light Microscopic Examination and Stains
Endomyocardial biopsy tissue that is going to be examined by
SPECIAL ISSUES routine light microscopy is embedded in paraffin and serially
Tissue Processing sectioned into 4 mm thick layers mounted on sequentially
The clinician performing the endomyocardial biopsy is respon- numbered glass slides. In cases of suspected myocarditis,
sible for obtaining adequate tissue for analysis and ensuring the most laboratories stain every third slide with hematoxylin
tissue is placed in the appropriate preservative. The number of and eosin for histomorphologic characterization. Two slides
specimens obtained depends on the clinical situation and are typically stained with Movat or elastic trichrome stain to
studies to be performed. Adequate diagnostic yield from visualize collagen and elastic tissue. In addition, many labo-
repeated biopsy must be balanced against the risk of the biopsy ratories will stain one slide for iron on specimens from men and
procedure. It is generally recommended that at least five sep- all postmenopausal women. Congo red staining is performed
arate specimens, each 1 to 2 mm3 in size, be obtained from more on a 10- to 15-mm section in all patients over the age of 50 to
than one region to minimize sampling error. For transplant identify cardiac amyloidosis. Table 13.5 summarizes the stains
rejection surveillance, the International Society of Heart and that are occasionally used in the evaluation of heart biopsy
Lung Transplantation requires a minimum of four biopsy samples depending on the clinical situation.
specimens, each with less than 50% of the sample being fibrous
tissue, thrombus, or other noninterpretable tissues, such as
those with crush artifact. The sensitivity of detecting transplant Molecular Studies
rejection approaches 98% when the pathologist reviews five Increasingly, molecular techniques are available which
adequate biopsy samples (76). improve the clinical utility of endomyocardial biopsy, above
To avoid contamination of the biopsy specimen, once the
bioptome forceps have been removed from the venous sheath Table 13.5 Special Stains for Endomyocardial Biopsy
and the jaws are opened, the sample should be gently extracted
with a sterile needle and placed immediately into preservative Stain Frequent indication
solution (10% neutral buffered formalin) (77,78). Fixative Methyl green pyronine Lymphocytes (myocarditis/
should be at room temperature to prevent contraction band allograft rejection)
artifacts (79). Excessive traction or crushing of the sample, as Movat pentachrome Connective tissue/vasculature
well as cutting a single larger sample into many, should be Masson elastic trichrome Connective tissue/vasculature
Prussian blue Iron
avoided because it may disrupt histologic architecture. Addi-
Sulfated Alcian blue Iron and amyloid
tional samples may be submitted for transmission electron Congo red Amyloid
microscopy to evaluate infiltrative diseases or anthracycline Ziehl-Neelsen Granulomas (acid fast bacilli)
toxicity (80). For transmission electron microscopy, pieces are Periodic acid Schiff Intramyocardial glycogen/
fixed in 4% glutaraldehyde at room temperature. One or more vasculture/fungi
samples may be frozen for molecular analysis, immunofluor- Gram Bacteria (endocarditis)
escence, immunohistochemistry, or viral genome analysis. To Gomori methenamine silver Fungi
prepare frozen specimens in the catheterization laboratory, Haematoxylin phloxine saffron Connective tissue
samples should be placed in embedding solution then snap- Methyl violet Amyloid
frozen in OCT-embedding medium or alternatively placed in Source: Reproduced from Ref. 79.
[2/7/010/9:43:7] [136–151]
and beyond the simple histopathologic and biochemical anal- CONCLUSIONS

ysis that has been available to this point. Advances in PCR Endomyocardial biopsy remains an integral mode of investiga-
techniques allow pathologists or investigators to determine tion for diagnosing many primary and secondary myocardial
whether or not the patient’s cardiomyopathy or myocarditis conditions. The modern approach to endomyocardial biopsy
is associated with a preexistent or ongoing viral infection (72). was introduced by Sakakibara and Konno in the early 1960s
Current PCR techniques can detect fewer than 10 gene copies of and then modified and popularized by the Stanford group in the
viral pathogens in an endomyocardial biopsy sample. For PCR early 1970s as a means to monitor graft rejection following
analysis to be considered reliable the biopsy sample must be cardiac transplantation. Since then, the right ventricular heart
rapidly and properly transported to the laboratory for analysis. biopsy procedure has gained acceptance as a useful investigative
Proper handling of the sample includes the use of pathogen- tool for nontransplant cardiac pathology. The indications for
free biopsy devices and storage vials and the transportation of endomyocardial biopsy have been outlined in a consensus state-
biopsy samples in RNA later solution on dry ice at room ment published in 2007 by the American Heart Association, the
temperature. American College of Cardiology, and the European Society of
Over the past 20 years, the use of PCR has increased our Cardiology. Indications for endomyocardial biopsy include post-
understanding about possible cardiotropic viruses in patients transplant rejection surveillance, investigation of infiltrative dis-
with unexplained cardiomyopathy. Numerous studies of orders of the myocardium, primary cardiomyopathies,
patients with myocarditis or dilated cardiomyopathy have myocarditis, endocardial fibrosis (as a way to help distinguish
reported a wide range of viruses including enteroviruses between constrictive and restrictive pathology), drug toxicity,
(most commonly Coxsackie B virus), adenoviruses, Parvovirus ventricular arrhythmias, unexplained heart failure in children,
B19, cytomegalovirus, influenza and respiratory syncytial and suspected cardiac neoplasia. Cardiac biopsy is extremely safe
virus, Ebstein-Barr virus, HIV, Hepatitis C virus, and human when performed by an experienced operator. Complication rates
herpes virus 6 (81–87). There are several limitations to the are reduced by appropriate patient selection, careful biopsy
widespread use of PCR in screening endomyocardial biopsy technique guided by fluoroscopy and ultrasound, as well as
samples for cardiotropic viruses. Currently available PCR- close patient monitoring. The role of endomyocardial biopsy
based viral isolation techniques remain labor intensive, costly, continues to evolve as novel molecular and genetic analyses
and lack standardization. Existing PCR screening methods are shed new light on heart muscle disorders.
also restricted to a limited number of predetermined candidate
viruses. Because the number of biopsy samples needed to attain REFERENCES
a clinically acceptable sensitivity for cardiotropic viruses is not 1. Weinberg M, Fell EH, Lynfield J. Diagnostic biopsy of the peri-
known, a positive PCR is diagnostic of viral infection, however, cardium and myocardium. AMA Arch Surg 1958; 76:825–829.
a negative PCR does not exclude viral infection. Lastly, and 2. Sutton DC, Sutton GC. Needle biopsy of the human ventricular
most importantly, presence of viral genomic material in biopsy myocardium: review of 54 consecutive cases. Am Heart J 1960; 60:
specimens does not prove causality of cardiomyopathy and 364–370.
currently does not change management strategy. 3. Shirey EK, Hawk WA, Mukerji D, et al. Percutaneous myocardial
Molecular studies can also be utilized to look for immune biopsy of the left ventricle. Experience in 198 patients. Circulation
markers, such as HLA upregulation and immune deposition to 1972; 46:112–122.
4. Timmis GC, Gordon S, Baron RH, et al. Percutaneous myocardial
identify those patients who have an autoimmune process that
biopsy. Am Heart J 1965; 70:499–504.
may be perpetuating ventricular dysfunction (74). Molecular 5. Bulloch RT, Murphy ML, Pearce MB. Intracardiac needle biopsy of
studies are not limited to diagnostic evaluation but have also the ventricular septum. Am J Cardiol 1965; 16:227–233.
been shown to have prognostic implications in patients with 6. Sakakibara S, Konno S. Endomyocardial biopsy. Jpn Heart J 1962;
new-onset idiopathic cardiomyopathy. Heidecker et al. dem- 3:537–543.
onstrated that microarray technology could be utilized to gen- 7. Caves PK, Schulz WP, Dong E Jr., et al. New instrument for
erate a transcriptomic signature (45 genes) from a single transvenous cardiac biopsy. Am J Cardiol 1974; 33:264–267.
endomyocardial biopsy, which could predict prognosis in 8. Caves PK, Stinson EB, Graham AF, et al. Percutaneous trans-
patients with new-onset heart failure of unclear etiology with venous endomyocardial biopsy. JAMA 1973; 225:288–291.
very high specificity (90%) (88). 9. Kawai C, Kitaura Y. New endomyocardial biopsy catheter for the
left ventricle. Am J Cardiol 1977; 40:63–65.
10. Richardson PJ. King’s endomyocardial bioptome. Lancet 1974; 1:
Future Directions 660–661.
11. Baughman K, Baim D. Endomyocardial biopsy. In: Baim D, ed.
The clinical utility of endomyocardial biopsy will no doubt be
Grossman’s Cardiac Catheterization, Angiography, and Intervention.
strengthened by the application of highly specific molecular 7th ed. Philadelphia: Lippincott Williams and Wilkins, 2006:395–412.
probes and microarray DNA technology used to look for viral 12. Baumgartner WA, Reitz BA, Kasper E, et al. Heart and Lung
genomic material and autoimmunity in heart biopsy specimens. Transplantation. 2nd ed. Philadelphia, PA: W. B. Saunders Com-
Our current understanding of idiopathic cardiomyopathies is pany, 2002.
that they are the consequence of a complex interplay between 13. Denys BG, Uretsky BF, Reddy PS, et al. An ultrasound method for
inflammatory, infectious, autoimmune, and genetic factors ulti- safe and rapid central venous access. N Engl J Med 1991; 324:566.
mately resulting in myocardial injury and remodeling. With the 14. Denys BG, Uretsky BF, Reddy PS. Ultrasound-assisted cannulation
improvement and increased availability of techniques to define of the internal jugular vein. A prospective comparison to the exter-
immune upregulation and viral persistence it is likely that nal landmark-guided technique. Circulation 1993; 87:1557–1562.
15. Anderson JL, Marshall HW. The femoral venous approach to
endomyocardial biopsies will redefine myocarditis and its appro-
endomyocardial biopsy: comparison with internal jugular and
priate treatment. Only through a detailed understanding of the transarterial approaches. Am J Cardiol 1984; 53:833–837.
pathobiology of idiopathic cardiomyopathies will we be able to 16. Brooksby IA, Jenkins BS, Coltart DJ, et al. Left-ventricular endo-
develop novel therapies targeted at these important disorders. myocardial biopsy. Lancet 1974; 2:1222–1225.
[2/7/010/9:43:7] [136–151]
17. Pierard L, El Allaf D, D’Orio V, et al. Two-dimensional echocardio- 37. Cooper LT, Zehr KJ. Biventricular assist device placement and
graphic guiding of endomyocardial biopsy. Chest 1984; 85:759–762. immunosuppression as therapy for necrotizing eosinophilic myo-
18. Miller LW, Labovitz AJ, McBride LA, et al. Echocardiography- carditis. Nat Clin Pract Cardiovasc Med 2005; 2:544–548.
guided endomyocardial biopsy. A 5-year experience. Circulation 38. Okura Y, Dec GW, Hare JM, et al. A clinical and histopathologic
1988; 78:III99–III102. comparison of cardiac sarcoidosis and idiopathic giant cell myo-
19. Mahrholdt H, Goedecke C, Wagner A, et al. Cardiovascular mag- carditis. J Am Coll Cardiol 2003; 41:322–329.
netic resonance assessment of human myocarditis: a comparison to 39. Baughman KL. Diagnosis of myocarditis: death of Dallas criteria.
histology and molecular pathology. Circulation 2004; 109:1250–1258. Circulation 2006; 113:593–595.
20. Amitai ME, Schnittger I, Popp RL, et al. Comparison of three- 40. Mason JW, O’Connell JB, Herskowitz A, et al. A clinical trial of
dimensional echocardiography to two-dimensional echocardiog- immunosuppressive therapy for myocarditis. The Myocarditis
raphy and fluoroscopy for monitoring of endomyocardial biopsy. Treatment Trial Investigators. N Engl J Med 1995; 333:269–275.
Am J Cardiol 2007; 99:864–866. 41. Dec GW, Fuster V. Idiopathic dilated cardiomyopathy. N Engl J
21. Rios B, Nihill MR, Mullins CE. Left ventricular endomyocardial Med 1994; 331:1564–1575.
biopsy in children with the transseptal long sheath technique. 42. Felker GM, Hu W, Hare JM, et al. The spectrum of dilated
Cathet Cardiovasc Diagn 1984; 10:417–423. cardiomyopathy. The Johns Hopkins experience with 1,278
22. Hunt SA, Haddad F. The changing face of heart transplantation. J patients. Medicine (Baltimore) 1999; 78:270–283.
Am Coll Cardiol 2008; 52:587–598. 43. Schoenfeld MH, Supple EW, Dec GW Jr., et al. Restrictive cardi-
23. Miller LW, Schlant RC, Kobashigawa J, et al. 24th Bethesda Con- omyopathy versus constrictive pericarditis: role of endomyocar-
ference: cardiac transplantation. Task Force 5: complications. J Am dial biopsy in avoiding unnecessary thoracotomy. Circulation
Coll Cardiol 1993; 22:41–54. 1987; 75:1012–1017.
24. Ballester M, Bordes R, Tazelaar HD, et al. Evaluation of biopsy 44. Arad M, Maron BJ, Gorham JM, et al. Glycogen storage diseases
classification for rejection: relation to detection of myocardial presenting as hypertrophic cardiomyopathy. N Engl J Med 2005;
damage by monoclonal antimyosin antibody imaging. J Am Coll 352:362–372.
Cardiol 1998; 31:1357–1361. 45. Falk RH. Diagnosis and management of the cardiac amyloidoses.
25. Deng MC, Eisen HJ, Mehra MR, et al. Noninvasive discrimination Circulation 2005; 112:2047–2060.
of rejection in cardiac allograft recipients using gene expression 46. Fowles RE, Mason JW. Endomyocardial biopsy. Ann Intern Med
profiling. Am J Transplant 2006; 6:150–160. 1982; 97:885–894.
26. Pham MX, Deng MC, Kfoury AG, et al. Molecular testing for long- 47. Sekiguchi M, Take M. World survey of catheter biopsy of the
term rejection surveillance in heart transplant recipients: design of heart. In: Sekiguchi M, Olsen E, eds. Cardiomyopathy: Clinical,
the Invasive Monitoring Attenuation Through Gene Expression Pathological and Theoretical Aspects. Baltimore: University Park
(IMAGE) trial. J Heart Lung Transplant 2007; 26:808–814. Press, 1980:217–225.
27. Stewart S, Winters GL, Fishbein MC, et al. Revision of the 1990 48. Deckers JW, Hare JM, Baughman KL. Complications of trans-
working formulation for the standardization of nomenclature in venous right ventricular endomyocardial biopsy in adult patients
the diagnosis of heart rejection. J Heart Lung Transplant 2005; with cardiomyopathy: a seven-year survey of 546 consecutive
24:1710–1720. diagnostic procedures in a tertiary referral center. J Am Coll
28. Cooper LT, Baughman KL, Feldman AM, et al. The role of Cardiol 1992; 19:43–47.
endomyocardial biopsy in the management of cardiovascular 49. Han J, Park Y, Lee H, et al. Complications of 2-D echocardiog-
disease: a scientific statement from the American Heart Associa- raphy guided transfemoral right ventricular endomyocardial
tion, the American College of Cardiology, and the European biopsy. J Korean Med Sci 2006; 21:989–994.
Society of Cardiology. Circulation 2007; 116:2216–2233. 50. Baraldi-Junkins C, Levin HR, Kasper EK, et al. Complications of
29. Heart Failure Society of America. HFSA 2006 Comprehensive endomyocardial biopsy in heart transplant patients. J Heart Lung
Heart Failure Practice Guideline. J Card Fail 2006; 12:e1–e2. Transplant 1993; 12:63–67.
30. Hunt SA. ACC/AHA 2005 guideline update for the diagnosis and 51. Holzmann M, Nicko A, Kuhl U, et al. Complication rate of right
management of chronic heart failure in the adult: a report of the ventricular endomyocardial biopsy via the femoral approach: a
American College of Cardiology/American Heart Association retrospective and prospective study analyzing 3048 diagnostic
Task Force on Practice Guidelines (Writing Committee to Update procedures over an 11-year period. Circulation 2008; 118:1722–1728.
the 2001 Guidelines for the Evaluation and Management of Heart 52. Pophal SG, Sigfusson G, Booth KL, et al. Complications of endo-
Failure). J Am Coll Cardiol 2005; 46:e1–e82. myocardial biopsy in children. J Am Coll Cardiol 1999; 34:
31. Swedberg K, Cleland J, Dargie H, et al. Guidelines for the diag- 2105–2110.
nosis and treatment of chronic heart failure: executive summary 53. Friedrich SP, Berman AD, Baim DS, et al. Myocardial perforation
(update 2005): the Task Force for the Diagnosis and Treatment of in the cardiac catheterization laboratory: incidence, presentation,
Chronic Heart Failure of the European Society of Cardiology. Eur diagnosis, and management. Cathet Cardiovasc Diagn 1994; 32:
Heart J 2005; 26:1115–1140. 99–107.
32. Felker GM, Thompson RE, Hare JM, et al. Underlying causes and 54. Ruesch S, Walder B, Tramer MR. Complications of central venous
long-term survival in patients with initially unexplained cardio- catheters: internal jugular versus subclavian access–a systematic
myopathy. N Engl J Med 2000; 342:1077–1084. review. Crit Care Med 2002; 30:454–460.
33. Amabile N, Fraisse A, Bouvenot J, et al. Outcome of acute 55. Hind D, Calvert N, McWilliams R, et al. Ultrasonic locating devices
fulminant myocarditis in children. Heart 2006; 92:1269–1273. for central venous cannulation: meta-analysis. BMJ 2003; 327:361.
34. Dec GW Jr., Palacios IF, Fallon JT, et al. Active myocarditis in the 56. Karakitsos D, Labropoulos N, De Groot E, et al. Real-time ultra-
spectrum of acute dilated cardiomyopathies. Clinical features, sound-guided catheterisation of the internal jugular vein: a pro-
histologic correlates, and clinical outcome. N Engl J Med 1985; spective comparison with the landmark technique in critical care
312:885–890. patients. Crit Care 2006; 10:R162.
35. McCarthy RE III, Boehmer JP, Hruban RH, et al. Long-term 57. Merrer J, De Jonghe B, Golliot F, et al. Complications of femoral
outcome of fulminant myocarditis as compared with acute (non- and subclavian venous catheterization in critically ill patients: a
fulminant) myocarditis. N Engl J Med 2000; 342:690–695. randomized controlled trial. JAMA 2001; 286:700–707.
36. Cooper LT Jr., Berry GJ, Shabetai R. Idiopathic giant-cell myocardi- 58. Kreher SK, Ulstad VK, Dick CD, et al. Frequent occurrence of
tis–natural history and treatment. Multicenter Giant Cell Myocardi- occult pulmonary embolism from venous sheaths during endo-
tis Study Group Investigators. N Engl J Med 1997; 336:1860–1866. myocardial biopsy. J Am Coll Cardiol 1992; 19:581–585.
[2/7/010/9:43:7] [136–151]
59. Bell RL, Atweh N, Ivy ME, Possenti P. Traumatic and iatrogenic 74. Wojnicz R, Nowalany-Kozielska E, Wojciechowska C, et al.
Horner syndrome: case reports and review of the literature. J Randomized, placebo-controlled study for immunosuppressive
Trauma 2001; 51:400–404. treatment of inflammatory dilated cardiomyopathy: two-year fol-
60. Martin-Hirsch DP, Newbegin CJ. Right vocal fold paralysis as a low-up results. Circulation 2001; 104:39–45.
result of central venous catheterization. J Laryngol Otol 1995; 75. Frustaci A, Chimenti C, Calabrese F, et al. Immunosuppressive
109:1107–1108. therapy for active lymphocytic myocarditis: virological and
61. Wong RC, Abrahams Z, Hanna M, et al. Tricuspid regurgitation immunologic profile of responders versus nonresponders. Circu-
after cardiac transplantation: an old problem revisited. J Heart lation 2003; 107:857–863.
Lung Transplant 2008; 27:247–252. 76. Billingham M. Pathology of heart transplantation. In: Solez K,
62. Braverman AC, Coplen SE, Mudge GH, et al. Ruptured chordae Racusen L, Billingham M, eds. Solid Organ Transplant Rejection:
tendineae of the tricuspid valve as a complication of endomyo- Mechanisms, Pathology and Diagnosis. New York: Marcel Dek-
cardial biopsy in heart transplant patients. Am J Cardiol 1990; ker, Inc., 1996:137–159.
66:111–113. 77. Virmani R, Burke A, Farb A, et al. Cardiovascular Pathology. 2nd
63. Mielniczuk L, Haddad H, Davies RA, et al. Tricuspid valve ed. Philadelphia: Saunders, 2001.
chordal tissue in endomyocardial biopsy specimens of patients 78. Veinot J, Ghadially F, Walley V. Light microscopy and ultra-
with significant tricuspid regurgitation. J Heart Lung Transplant structure of the blood vessel and heart. In: Silver M, Gotlieb A,
2005; 24:1586–1590. Schoen F, eds. Cardiovascular Pathology. 3rd ed. New York:
64. Nguyen V, Cantarovich M, Cecere R, et al. Tricuspid regurgitation Churchill Livingstone, 2001:30–53.
after cardiac transplantation: how many biopsies are too many? J 79. Cunningham KS, Veinot JP, Butany J. An approach to endomyo-
Heart Lung Transplant 2005; 24:S227–S231. cardial biopsy interpretation. J Clin Pathol 2006; 59:121–129.
65. Sandhu JS, Uretsky BF, Zerbe TR, et al. Coronary artery fistula in 80. Billingham ME, Mason JW, Bristow MR, et al. Anthracycline
the heart transplant patient. A potential complication of endo- cardiomyopathy monitored by morphologic changes. Cancer
myocardial biopsy. Circulation 1989; 79:350–356. Treat Rep 1978; 62:865–872.
66. Hauck AJ, Kearney DL, Edwards WD. Evaluation of postmortem 81. Bowles NE, Ni J, Kearney DL, et al. Detection of viruses in
endomyocardial biopsy specimens from 38 patients with lympho- myocardial tissues by polymerase chain reaction. evidence of
cytic myocarditis: implications for role of sampling error. Mayo adenovirus as a common cause of myocarditis in children and
Clin Proc 1989; 64:1235–1245. adults. J Am Coll Cardiol 2003; 42:466–472.
67. Chow LH, Radio SJ, Sears TD, et al. Insensitivity of right ven- 82. Grasso M, Arbustini E, Silini E, et al. Search for Coxsackievirus B3
tricular endomyocardial biopsy in the diagnosis of myocarditis. J RNA in idiopathic dilated cardiomyopathy using gene amplifica-
Am Coll Cardiol 1989; 14:915–920. tion by polymerase chain reaction. Am J Cardiol 1992; 69:658–664.
68. Winters G, Schoen F. Pathology of cardiac transplantation. In: 83. Jin O, Sole MJ, Butany JW, et al. Detection of enterovirus RNA in
Silver M, Gotlieb A, Schoen F, eds. Cardiovascular Pathology. myocardial biopsies from patients with myocarditis and cardio-
Philadelphia: Churchill Livingstone, 2001:725–759. myopathy using gene amplification by polymerase chain reaction.
69. Aretz HT, Billingham ME, Edwards WD, et al. Myocarditis. A Circulation 1990; 82:8–16.
histopathologic definition and classification. Am J Cardiovasc 84. Kuhl U, Pauschinger M, Noutsias M, et al. High prevalence of
Pathol 1987; 1:3–14. viral genomes and multiple viral infections in the myocardium of
70. Martin AB, Webber S, Fricker FJ, et al. Acute myocarditis. Rapid adults with “idiopathic” left ventricular dysfunction. Circulation
diagnosis by PCR in children. Circulation 1994; 90:330–339. 2005; 111:887–893.
71. Pauschinger M, Bowles NE, Fuentes-Garcia FJ, et al. Detection 85. Matsumori A. Hepatitis C virus infection and cardiomyopathies.
of adenoviral genome in the myocardium of adult patients with Circ Res 2005; 96:144–147.
idiopathic left ventricular dysfunction. Circulation 1999; 99: 86. Muir P, Nicholson F, Jhetam M, et al. Rapid diagnosis of enter-
1348–1354. ovirus infection by magnetic bead extraction and polymerase
72. Pauschinger M, Doerner A, Kuehl U, et al. Enteroviral RNA chain reaction detection of enterovirus RNA in clinical specimens.
replication in the myocardium of patients with left ventricular J Clin Microbiol 1993; 31:31–38.
dysfunction and clinically suspected myocarditis. Circulation 87. Tschope C, Bock CT, Kasner M, et al. High prevalence of cardiac
1999; 99:889–895. parvovirus B19 infection in patients with isolated left ventricular
73. Why HJ, Meany BT, Richardson PJ, et al. Clinical and prognostic diastolic dysfunction. Circulation 2005; 111:879–886.
significance of detection of enteroviral RNA in the myocardium of 88. Heidecker B, Kasper EK, Wittstein IS, et al. Transcriptomic bio-
patients with myocarditis or dilated cardiomyopathy. Circulation markers for individual risk assessment in new-onset heart failure.
1994; 89:2582–2589. Circulation 2008; 118:238–246.
[24/6/010/19:44:35] [152–156]
14
Pericardiocentesis
Carl L. Tommaso
INTRODUCTION Pericardial effusions can be inflammatory, uremic, infectious,

Pericardiocentesis is the transcutaneous drainage of fluid from or oncologic in nature or due to congestive heart failure (1).
the pericardium. It may be performed for diagnostic or thera- Therapeutic pericardiocentesis is performed for the relief of
peutic indications. Pericardiocentesis may be performed on an cardiac tamponade. Tamponade may be due to any of the
elective or emergent basis and when done emergently may be above etiologies, but is most commonly due to trauma such
life saving. It may be done to withdraw fluid from the peri- as perforation of a cardiac structure during an intervention (2)
cardial space for diagnostic testing or may involve insertion of or closed chest trauma.
a drain for relieving the pericardial effusion over hours to days.
APPROACH
ANATOMIC CONSIDERATIONS There are several approaches to the performance of a pericar-
The pericardium consists of two layers: the visceral pericardiocentesis regarding guidance, technique, and location. The
dium is adherent to the epicardium, and the parietal pericar- most common approach is echocardiographically guided (3).
dium has attachments to the diaphragm, the sternum, and the Prior to the procedure an echocardiogram is obtained to ascer-
anterior mediastinum. The parietal pericardium is normally tain the size and location of the effusion. A small or non-
about 1 to 2 mm thick and holds the heart in a relatively fixed compressive effusion may not be necessary to drain under
position during respiration and changes in body position. The most circumstances.
two layers are continuous and form a “sac” surrounding most The preechocardiogram technique of pericardiocentesis
of the heart except for the left atrium that is mostly extraper- used the electrocardiogram (EKG) as the marker (4). An alliga-
icardial. The pericardial reflection includes the origins of the tor clip was connected from the needle to a unipolar EKG lead
vena cava and the origins of the great vessels. The phrenic and monitoring of the EKG was done as the needle was
nerves are the only nonvascular structures included within the inserted. If the needle touched the myocardium, an injury
pericardium. current (ST segment elevation) would be present on the EKG
The vena cava and right ventricle are the most anterior indicating the needle had been placed too far.
cardiac structures and attempted pericardiocentesis may risk Most effusions are free flowing within the pericardial
perforating or entering either of these structures. The left ante- space and may be accessed from an anterior approach, but the
rior descending coronary artery is the most anterior of the few loculated effusions that occur will need to be approached by
coronary arteries, and while unlikely, can be punctured during a different technique (described below) (5). Some institutions
pericardiocentesis. The internal thoracic arteries are located on have substituted computed tomography (CT) scans (6) for the
the posterior surface of the chest wall in the midclavicular line. echocardiogram, and while this may be as effective, the porta-
Although not usually in the path of a pericardiocentesis needle, bility and ease of use make the echocardiogram the imaging
during attempts at a lateral loculated effusion they could be in modality of choice.
the pathway. Once the size and location of the effusion has been deter-
Within the normal pericardial sac is approximately 50 mL mined, echocardiogram may be used to visualize the needle to
of serous fluid. The normal pericardium is relatively inelastic ascertain its location in the pericardial space.
and small amounts of additional fluid (<100 mL) may com- In an elective procedure, a right heart catheterization
promise filling of cardiac chambers (cardiac tamponade). With with simultaneous measurement of pulmonary capillary
chronic fluid accumulation, however, the pericardium may wedge (PCW) pressure, right atrial (RA) pressure, and intra-
progressively stretch, become elastic, and accommodate large pericardial pressure is an elegant method of diagnosis of
amounts of fluid without hemodynamic alteration. pericardial tamponade, but it is usually not necessary.
When pericardial effusions occur the fluid may be dis- In an emergency, such as a vascular or chamber perfora-
tributed throughout the pericardial space or may be loculated, tion during a cardiac intervention with free-flowing blood
depending on the etiology of the effusion (Table 14.1) and prior (contrast) into the pericardium and resultant hemodynamic
pericardial disease or surgery. deterioration, no imaging is necessary and pericardiocentesis
should be performed as quickly as possible.
To bring the effusion inferior and closer to the chest wall,
INDICATIONS especially in free-flowing effusions, the elective procedure is best
Pericardiocentesis may be performed for diagnostic or thera- done with the patient sitting up at a 308 to 458 angle (Fig. 14.1). In
peutic reasons. The most common diagnostic indication is to loculated effusions the patient should be positioned to bring the
discern the etiology of accumulated pericardial fluid, usually effusion closest to the chest wall and may require the patient
in a patient with a chronic effusion of uncertain etiology. being placed into an extreme lateral decubitus position (7).
[24/6/010/19:44:35] [152–156]
PERICARDIOCENTESIS 153
Table 14.1 Causes of Pericardial Effusions

I. Collagen vascular disease
II. Iatrogenic
1. Cardiac surgery
2. CVP catheter insertion
3. Endomyocardial biopsy
4. Pacemaker lead (temporary or permanent)
5. Percutaneous coronary intervention
a. Cutting balloon
b. Associated with Glycoprotein IIb/IIIa inhibitor use
c. Guidewire perforation
d. Saphenous vein graft rupture
6. Radiofrequency ablation
7. Trans-septal puncture
8. Valvuloplasty
9. Ventriculogram
III. Infection
IV. Malignancy
V. Trauma
VI. Uremia
Figure 14.2 Locations for needle insertion for pericardiocentesis.

Location 1 is just left (patient’s left) of the xiphoid and is the most
common location used. Alternatives include location 2, to the right of
the xiphoid, and location 3, below the xiphoid. Locations 4 and 5 are
in the fifth intercostal space on either side of the sternum, and 6 and
7 are in the midclavicular line as the insertion site for lateral
loculated effusions.
Figure 14.1 Patient in semiupright (30–458) position in preparation

for pericardiocentesis.
Local anesthesia is administered as a skin wheal with a

21-gauge needle and then deeper anesthesia is infused with a
1.5-in 18-gauge needle. On occasion this needle will enter the
pericardium as signaled by straw-colored fluid return. This
needle can be used as a guide and the pericardiocentesis needle
inserted alongside it.
The typical location for skin entry is at the left xiphoid
notch aiming toward the left midclavicle (Figs. 14.2–14.5). The
needle is initially advanced horizontally and once under the
notch it is angled upward toward the clavicle. The position of
entry, patient positioning, and puncture site may be altered
depending on the location of the effusion.
Other sites of entry include the fifth or sixth intercostal
space at the left or right of the sternum. Using this approach,
the patient should be in the 308 to 458 upright position and the
needle angled to the mid clavicle. The potential advantage to
this approach is that the puncture site is already above the
diaphragm, so abdominal contents/organs are avoided.
The apical approach is also occasionally used. Since the Figure 14.3 Bony landmarks superimposed on the trunk.
point of maximal impact will not be present in the setting of a
[24/6/010/19:44:35] [152–156]
determined by the echocardiogram (8). Since the internal tho-

racic arteries lie in the midclavicular line they may be punc-
tured during a midclavicular line approach. Upon insertion of
the needle, pulsatile red blood may signal puncture of the
internal thoracic artery.
Since the pericardium is thick, resistance may be felt and
a “pop” noted upon entering the pericardial space. Depending
on the etiology of the pericardial fluid it may be serous or
frankly bloody. The old adage, “blood in the pericardium will
not clot,” can no longer be trusted to distinguish pericardial
blood from intracardiac blood because patients may be on
potent antiplatelet and/or anticoagulant medications. Pressure
measurement can be helpful, but the pressures in the right
atrium and vena cava may be identical to pericardial pressure,
although a right ventricular pressure will be obvious (9).
If in doubt of the needle location injection of nonionic
contrast with x-ray visualization or saline administration with
echocardiographic visualization will be helpful in ensuring the
presence in the pericardium (Fig. 14.6). Once the needle posi-
tion has been ascertained measurement of pressure to compare
with the RA pressure can be made to confirm the presence of
tamponade. Withdrawal of as little as 50 to 100 mL of fluid may
be adequate to relieve the pressure within the pericardium and
restore hemodynamics in the presence of cardiac tamponade.
Figure 14.4 Direction of the needle, the broad arrow is the initial
insertion, “walking above the diaphragm” and the narrow arrow
indicates aiming toward the left midclavicle.
Figure 14.5 Bony structures of the thorax with the location of the
cardiac structure behind to demonstrate the cardiac structures that
can be punctured or lacerated during pericardiocentesis.
pericardial effusion, the apex is identified by echocardiogram,

and the needle inserted toward the largest accumulation of Figure 14.6 Depiction by echocardiogram (A) when needle is
fluid identified by the echocardiogram. The approach to lateral placed properly in pericardium and saline is injected and (B) when
loculated pericardial effusions is also through puncture at the needle is placed in the right ventricle and saline is injected. The
midclavicular line. The key to puncturing a loculated lateral marked area in frame A represents pericardial space and in frame
effusion, whether right or left sided, is to position the patient to B represents right ventricular cavity.
bring the effusion as close as possible to the chest wall as
[24/6/010/19:44:35] [152–156]
PERICARDIOCENTESIS 155
If the pericardiocentesis is being performed only for diag- Cardiac tamponade has nonspecific physical findings,
nostic purposes in a large effusion, adequate specimens may be including paradoxical pulse and inspiratory decrease in sys-
withdrawn through the needle, however, it is probably best to temic venous pressure. Distant heart sounds, when noted with
exchange the needle for a soft tip catheter or introducer sheath. the other physical findings, are a further clue; however, this
When a drain is to be placed a 0.035-in guidewire is inserted constellation of findings may be present in other disease states
through the needle and confirmed in the pericardium by fluoros- including severe obstructive lung disease. Hemodynamic com-
copy. A 6- or 7-Fr tapered dilator is then placed for ease of promise includes hypotension (or a relative decrease from
inserting the drain. This dilator can be used to withdraw further baseline pressure) and tachycardia (which may be masked by
fluid in a diagnostic-only tap. A drain, usually a pig tail–shaped or the presence of b-blockers). As right-sided cardiac chamber
straight catheter with multiple side holes is then inserted. The filling decreases, stroke volume also decreases. The heart rate
optimum location for this catheter is inferior or posterior to the increases in an effort to maintain cardiac output. The most
heart. To ascertain position, nonionic contrast may be injected. common noninvasive method of determining the presence of
The drainage catheter is then connected to a drainage bag tamponade is by echocardiography. Right heart chamber col-
via a plastic tube (10). Hand withdrawal of fluid may be lapse in the presence of pericardial effusion during diastole is
necessary to remove enough fluid to decompress the tampo- sensitive for the presence of tamponade (11,12).
nade. Once the drainage catheter is in place and hemodynamics Asymptomatic pericardial effusions can be suspected by
have been restored, the catheter is suture in place and gravity an abnormal chest x-ray with a typical water-bottle configura-
or low negative pressure drainage begun. tion of the cardiac silhouette. The chest x-ray findings can be
The pericardial fluid is usually sent for cell count, differ- confirmed by echocardiography or chest CT or MR scans, or the
ential, chemistries (total protein, pH, LDH, glucose, etc.), serol- pericardial effusion may be first noted on these modalities.
ogies, culture, and cytology. The most accurate test for the presence of cardiac tam-
Depending on the situation the pericardial drain is left in ponade is the hemodynamic assessment of intracardiac pres-
place for 12 hours or longer, and once the pericardial effusion has sures simultaneous with the measurement of intrapericardial
been drained as confirmed by echocardiography it can be removed. pressure. In the presence of tamponade there will be diastolic
equilibration of the pressures equal to the intrapericardial
pressure.
EQUIPMENT The amount of fluid necessary to cause tamponade is
In an elective pericardiocentesis, an echocardiogram at the variable and depends on the elasticity/stiffness of the parietal
bedside is necessary for quantitating and localizing the effusion pericardium. When pericardial fluid accumulates rapidly in a
as well as determining needle placement. stiff pericardium, the amount of fluid to cause tamponade will
As noted above the best position for most pericardiocent- be relatively small. In contrast, when accumulation of fluid is
esis procedures is sitting up at approximately 308 to 458 and slow (i.e., over weeks to months) and the pericardium has time
this is best accomplished by a wedge placed behind the patient, to stretch, even very large amounts of fluid will not cause
or if done in bed, elevating the back rest to 458 or greater as hemodynamic compromise.
such a radio-transparent wedge is necessary. To monitor needle Given the mechanical nature of tamponade, the only
position an EKG or cardiac monitor with a unipolar lead is treatment options are pericardiocentesis or surgical drainage.
necessary. Other equipment is usually provided in a commer- Advantages of pericardiocentesis over surgical drainage are the
cially available prepackaged pericardiocentesis kit and the ability to perform it quickly and at bedside if necessary.
contents are outlined in Table 14.2. Diuretics may worsen the situation by reducing the intracardiac
pressures and hastening filling of the heart.
In cardiac tamponade stroke volume is reduced and
CLINICAL ASPECTS maintenance of blood pressure and cardiac output may be
No specific symptoms are associated with pericardial effusions. achieved by an increase in heart rate. A delay in performing
Chest pain, dyspnea at rest or with exertion, pleuritic pain, or pericardiocentesis may necessitate increasing fluid volume and
palpitations alone or together may be present. heart rate with isoproterenol or dopamine for the short-term
support of blood pressure.
Table 14.2 Equipment (Much of This Is Packaged Together in Typically, once a pericardial drain has been placed, the
Disposable Pericardiocentesis Tray) amount of pericardial fluid is monitored in the collection sys-
l 0.038-in guidewire tem. In addition, serial echocardiograms may be performed. If
l 18-gauge 1.5-in needle fluid continues to drain but no reduction in the size of the
l 21-gauge skin needle effusion is noted, continued accumulation of fluid (bleeding)
l 30–458 wedge may make surgical intervention necessary.
l 4–5 in long 18-gauge needle
l 6-Fr dilator
l 6-Fr pig tail or specialized pericardial drain
l Alligator clip CONTRAINDICATIONS
l Echocardiogram machine The contraindications to pericardiocentesis include the follow-
l Electrocardiogram machine or monitor ing:
l Lidocaine
l Skin cleansing agent 1. An inadequate amount of pericardial fluid, particularly in
l Sterile drapes(s) situations of acute/chronic pericarditis, may result in rapid
l Stopcock, tubing and drainage bag resolution of the effusion, and the performance of an
l Tubes for chemistry and culture echocardiogram immediately prior to the pericardiocent-
esis is necessary.
[24/6/010/19:44:35] [152–156]
2. Coagulopathy should be corrected prior to performing an REFERENCES

elective diagnostic pericardiocentesis to reduce iatrogenic 1. Vaitkus PT, Herrmann HC, LeWinter MM. Treatment of malig-
bleeding. However, in the presence of tamponade, peri- nant pericardial effusion. JAMA 1994; 272:59–64.
cardiocentesis even in the presence of coagulopathy may 2. Tsang TS, Freeman WK, Barnes ME, et al. Rescue echocardio-
be life saving. graphically guided pericardiocentesis for cardiac perforation com-
plicating catheter-based procedures. The Mayo Clinic experience.
3. Loculated lateral or posterior effusions may be difficult or
J Am Coll Cardiol 1998; 32:1345–1350.
impossible to reach with a needle, and surgical drainage 3. Callahan JA, Seward JB, Nishimura RA, et al. Two-dimensional
should be considered. echocardiographically guided pericardiocentesis: experience in
117 consecutive patients. Am J Cardiol 1985; 55:476–479.
4. Bishop LH Jr, Estes EH Jr, McIntosh HD. The electrocardiogram as a
COMPLICATIONS safeguard in pericardiocentesis. J Am Med Assoc 1956; 162:264–265.
5. Callahan JA, Seward JB, Tajik AJ. Cardiac tamponade: pericardio-
One complication of pericardiocentesis is laceration of cardiac
centesis directed by two-dimensional echocardiography. Mayo
or noncardiac structures. Because the right ventricle and vena Clin Proc 1985; 60:344–347.
cava lie anterior they are the most common cardiac structures 6. Duvernoy O, Magnusson A. CT-guided pericardiocentesis. Acta
to be inadvertently punctured; however, coronary arteries, Radiol 1996; 37:775–778.
coronary veins, and the internal thoracic artery may also be 7. Tsang TS, Freeman WK, Sinak LJ, et al. Echocardiographically
lacerated. Serous colored fluid that quickly turns bloody may guided pericardiocentesis: evolution and state-of-the-art tech-
be an indication of laceration of a cardiac structure. While rare, nique. Mayo Clin Proc 1998; 73:647–652.
noncardiac structures including the liver, spleen, and stomach 8. Tsang TS, Enriquez-Sarano M, Freeman WK, et al. Consecutive
can be lacerated. The easiest way to avoid inadvertent puncture 1127 therapeutic echocardiographically guided pericardiocente-
of these structures is to carefully select the path of the peri- ses: clinical profile, practice patterns, and outcomes spanning 21
years. Mayo Clin Proc 2002; 77:429–436.
cardiocentesis needle. Inject iodinated contrast or saline with
9. Kilpatrick ZM, Chapman CB. On pericardiocentesis. Am J Cardiol
x-ray or echocardiographic imaging to demonstrate the location 1965; 16:722–728.
of a puncture (13). 10. Kopecky SL, Callahan JA, Tajik AJ, et al. Percutaneous pericardial
Occasionally, congestive heart failure or pulmonary catheter drainage: report of 42 consecutive cases. Am J Cardiol
edema may result after drainage of a large pericardial effusion. 1986; 58:633–635.
Echocardiography may be necessary to differentiate these 11. Armstrong WF, Schilt BF, Helper DJ, et al. Diastolic collapse of the
symptoms from reaccumulation of pericardial fluid. A rare right ventricle with cardiac tamponade: an echocardiographic
phenomenon after relief of a pericardial effusion is biventricu- study. Circulation 1982; 65:1491–1496.
lar decompensation. The etiology of this is uncertain. It is 12. Kronzon I, Cohen ML, Winer HE. Diastolic atrial compression: a
usually transient, but may require pressors of mechanical sup- sensitive echocardiographic sign of cardiac tamponade. J Am Coll
Cardiol 1983; 2:770–775.
port for a short period of time.
13. Duvernoy O, Borowiec J, Helmius G, et al. Complications of
Dysrhythmias can occur during pericardiocentesis, and percutaneous pericardiocentesis under fluoroscopic guidance.
may be due to irritation from the needle or catheter on the Acta Radiol 1992; 33:309–313.
epicardium.
[7/7/010/21:13:41] [157–172]
15
Catheterization of the cardiac venous system

John C. Gurley
INTRODUCTION ANATOMY OF THE CARDIAC VEINS

For decades, the cardiac venous system was overlooked. Most Terminology of the Cardiac Venous System
clinicians considered the coronary sinus (CS) an obscure and Descriptions of the cardiac venous system can be confusing.
inaccessible structure with little purpose other than passive What is the difference between the great cardiac vein (GCV)
venous drainage. But in recent years the CS has emerged as a and the CS? What name should be used to describe the location of
vital pathway to the left heart that enables mainstream therapies a pacing lead, or to indicate a vein that is unsuitable because of
such as cardiac resynchronization and emerging therapies such diaphragmatic stimulation? These questions arise because most
as mitral valve repair, myocardial perfusion and gene therapy. clinicians learn to perform catheter-based procedures without the
Interventional cardiologists and electrophysiologists are now benefit of a structured, anatomical background. Therefore, this
expected to have a good working knowledge of cardiac venous section begins with an overview of the cardiac venous terminol-
anatomy, as well as the skill to safely catheterize the CS and its ogy (Fig. 15.1). We will see that the distinction between the CS
branches. This chapter will provide a practical review of the and epicardial veins is not arbitrary. A correct use of terminology
anatomy and physiology of the CS, the techniques of catheter- indicates a precise understanding of anatomy.
ization, and the potential applications of CS catheterization. The myocardium is drained by two interconnecting
After studying this chapter, the reader will understand venous systems, the Thebesian system and the epicardial sys-
the following: tem. The epicardial system can be subdivided into the CS and
branch epicardial veins (Table 15.2). The branch veins can be
1. Why catheterization of the CS and cardiac veins is
further grouped as veins that are always present or veins that
important
are variably present (Table 15.3).
2. How to use accurate terminology to describe the cardiac
venous system
3. What is the significance of Thebesian veins Thebesian Veins and the Intramyocardial
4. Why the CS should be considered the fifth chamber of the Collateral Network
heart rather than a vein Thebesian veins are small, valveless channels that drain the
5. How to identify the major epicardial veins by name and myocardium directly into the nearest cardiac chamber. They
location are found in all four chambers and produce a small, physio-
6. How to recognize the major vein-to-vein collateral loops logic right-to-left shunt of about 0.3% of aortic flow (2,3).
7. How to localize the CS ostium during catheterization The combination of vein-to-chamber, capillary-to-chamber,
8. How to catheterize the cardiac venous system safely and and artery-to chamber channels creates a rich network that
efficiently interconnects coronary arteries, veins, and cardiac chambers.
9. How to avoid complications of CS catheterization The Thebesian network provides an alternate route for
10. How to perform and interpret balloon-occlusion veno- drainage of the myocardium when epicardial veins become
graphy obstructed. This explains, in part, why balloon-occlusion
11. How to recognize abnormalities of the CS venography does not impair venous outflow, and why retro-
12. Why emerging therapies will require knowledge of cardiac grade contrast injections are remarkably well tolerated, even in
venous anatomy and skill at catheterization patients with advanced heart failure (4). Thebesian channels
facilitate vein-to-vein collateralization that develops when
major epicardial veins are obstructed by pacing leads or
Why Catheterize the CS? trauma. Thebesian veins may also play a nutritive role. The
Cardiac resynchronization therapy (CRT) has revolutionized relative immunity of atrial and right ventricular free walls to
the management of patients with heart failure. The dramatic ischemic infarction has been attributed to protective collateral-
improvements in ventricular function, quality of life and mor- ization by the Thebesian network.
tality have provided most of the incentive to understand the CS Thebesian drainage is sometimes visualized as plumes of
and to develop the tools and techniques that facilitate catheter- contrast entering the right and left ventricles (LV) during cor-
ization (1). But even with the best tools, cardiac resynchroniza- onary arteriography. Thebesian drainage is also seen during
tion is operator-dependent, requiring the safe delivery of a left alcohol septal ablation for hypertrophic cardiomyopathy, when
ventricular pacing lead to a location that is anatomically desir- echo contrast material injected into a septal perforator artery
able, stable, and free from phrenic nerve stimulation. Effective flows directly into the ventricles. Occasionally, a large Thebe-
CRT demands a thorough understanding of cardiac venous sian vein is inadvertently cannulated by an end-hole catheter
anatomy as well as some specific catheter techniques. Other (Fig. 15.2) (5). In each case, Thebesian drainage must be recog-
reasons to catheterize the CS are listed in Table 15.1. nized as normal anatomy and not a vascular malformation.
[7/7/010/21:13:41] [157–172]
Table 15.1 Reasons to Catheterize the Coronary Sinus

l Cardiac resynchronization
l Percutaneous mitral annuloplasty
l Arrhythmia management
l Myocardial drug delivery
l Gene and cell therapy
l Retrograde myocardial perfusion
Figure 15.2 Thebesian veins. Contrast injection during left ventri-

culography opacifies the cardiac venous system. The catheter tip has
inadvertently entered a Thebesian vein. Thebesian channels fill an
extensive network of interconnecting venous collaterals (black
arrows) that drain to the epicardial veins (arrowheads) and the cor-
onary sinus (white arrow). The sponge-like intramyocardial network
resembles the polymer cast in Figure 3.2.1. Source: From Ref. 5.
Figure 15.1 External view of the cardiac venous system. Anterior

view illustrating the AIV and the GCV (left). Note how the AIV makes
a 908 bend as it transitions from the anterior interventricular groove
to the atrioventricular groove; the name changes from AIV to GCV at The Thebesian network is essential to retrograde myo-
this point. Posterior view illustrating the continuation of the GCV into
cardial perfusion, drug delivery and gene therapy.
the CS (right). Also shown are the PIV or MCV, a PCV and the OVM.
Note how the GCV crosses marginal branches of the circumflex
artery. Abbreviations: AIV, anterior interventricular vein; PIV, pos- CS: the Fifth Chamber
terior interventricular vein; GCV, great cardiac vein; SCV, superior It is a common misconception that the CS is a simply a large
vena cava; CS, coronary sinus; MCV, middle cardiac vein; PCV,
vein formed by the confluence of epicardial veins. In fact, the
posterior cardiac vein; OVM, vein of Marshall. Source: From Netter
CS is not a vein at all. It is a small, contractile chamber complete
collection (annotations added).
with myocardium, two valves, and an electrical conduction
system (Fig. 15.3). The CS is very consistent in its structure and
position relative to the rest of the heart (6,7).
Externally, the CS lies in the atrioventricular groove
Table 15.2 The Cardiac Venous System behind the left atrium (LA), just above the usual location of
the circumflex artery. It lies completely outside the LA, being
separated by connective tissue and fat (8,9).
Internally, the CS has three distinct layers: endocardium,
myocardium and epicardium just like the other chambers of the
heart. The lumen of the CS is trabeculated, like other cardiac
chambers (and unlike veins) (10). The muscular wall consists of
striated cardiac muscle with intercalated disks, not the vascular
smooth muscle of veins (11), and the CS contracts—sometimes
vigorously—in synchrony with the atria. Normal contractions
can be observed by transthoracic or intracardiac ultrasound, or
during the venous phase of coronary arteriography (Fig. 15.4)
(12,13). It is important to understand normal contractility,
which can be mistaken for strictures and lead to unnecessary
balloon dilatation or abandonment of catheter procedures.
Table 15.3 Epicardial Veins

Consistent veins Variable veins
Anterior interventricular Posterior interventricular vein Vein of Posterior LV Lateral LV Anterolateral LV
vein (middle cardiac vein) Marshall veins veins veins
Abbreviation: LV, left ventricle.
[7/7/010/21:13:41] [157–172]
CATHETERIZATION OF THE CARDIAC VENOUS SYSTEM 159
Contractions of the CS can be forceful enough to eject balloon

occlusion catheters during venography.
Rhythmic contractions of the CS are absent in atrial
fibrillation, indicating that activation of the CS myocardium is
dependent on atrial inputs. These inputs consist of electrically
conductive muscle bands that connect the CS to both the right
atrium (RA) and the LA (14). These connections “bridge” the
atria and may help synchronize normal atrial contractions.
They can also allow the CS to participate in reentrant atrial
arrhythmias by forming left posterior or posterior paraseptal
accessory pathways that are difficult to map and ablate. The CS
receives additional inputs from the vein of Marshall (OVM),
which has been shown to be an autonomically innervated,
electrically active trigger for atrial fibrillation. Finally, cells
within the CS muscle cells may be capable of intrinsic automa-
ticity, as with other chambers (15).
The CS has two valves. The valve of Vieussens lies at the
junction of the GCV and the CS (Fig. 15.5). It is present in 90% of
adult hearts, has one or two cusps, and is variably competent
(16,17). It is important to recognize the valve of Vieussens because
it can be mistaken for a stenosis or occlusion. The valve of
Figure 15.3 Schematic diagram of the CS. The CS is located Vieussens marks the dividing line between the contractile CS
between the Thebesian valve at the ostium (T) and the valve of and the noncontractile GCV, which is an important landmark
Vieussens (V). It is defined by its muscular, contractile wall. The
during balloon-occlusion venography. The valve of Vieussens is
GCV (G) is a thin-walled venous structure that is separated from the
often associated with a mild, ring-like, less-distensible narrow-
CS by the valve of Vieussens. The GCV is the continuation of the
anterior interventricular vein; it is invariably present in this location. ing—essentially the annulus of the valve of Vieussens—which
The vein of Marshall (M) enters the CS just downstream from the can be exploited during balloon-occlusion venography.
valve of Vieussens. Branch veins can enter the CS or the GCV, and The Thebesian valve is a flap-like structure that covers the
they may contain venous valves. The middle cardiac vein (1) is also CS ostium to a variable degree (18). It is present in at least two
known as the posterior interventricular vein; it consistently enters thirds or adults, and can range from a scant remnant to a
the CS near its ostium. Other cardiac veins that are variably present complete covering (19,20). It can also exist as a mesh-like,
include a posterior vein (2) and a lateral vein (3). Abbreviations: multifenestrated membrane. Regardless of its appearance, the
GCV, great cardiac vein; CS, coronary sinus. Thebesian valve is always attached posteriorly (open anteriorly),
a feature that helps direct CS flow toward the tricuspid valve.
Each of the major veins that empty into the CS can also
contain valves near their openings. These valves can impede
the passage of catheters and pacing leads.
Figure 15.4 Normal contractility of the CS observed during coronary arteriography. (A) Arterial phase. (B) Venous phase in atrial diastole.
The junction of the CS and GCV is indicated by a subtle annular narrowing (arrow). (C) Venous phase during atrial systole. The distinction
between contractile CS and noncontractile GCV is apparent (arrow). A large posterolateral vein draining to the GCV provides a position
reference. Abbreviations: GCV, great cardiac vein; CS, coronary sinus.
[7/7/010/21:13:41] [157–172]
Figure 15.5 Valves of the CS. Contrast venography demonstrates a Thebesian valve at the mouth of the CS (small arrows) and a bileaflet
valve of Vieussens (open arrow). The muscular, contractile portion of the CS lies between these two valves. The oblique vein of the left atrium
(vein of Marshall) enters just downstream from the valve of Vieussens. Abbreviation: CS, coronary sinus.
In summary, it is helpful to think of the CS as the fifth follows the posterior interventricular groove from apex to base,
chamber of the heart. An appreciation of its structure and func- emptying directly into the CS near its ostium. The PIV is also
tion provides the foundation for safe and efficient catheterization. known as the middle cardiac vein (MCV).
The AIV and PIV connect at the apex, forming a semicircle
that establishes the plane of the interventricular septum.
Epicardial Veins The GCV and the CS form another semicircle that establishes
The epicardial veins carry most of the venous return from the
the plane of the mitral annulus. The intersection of these two
LV to the CS, which empties into the RA. They provide a
semicircles establishes the outline of the LV and provides a frame
transvenous path to the LV that can be utilized for cardiac
of reference for orienting all remaining veins (Fig. 15.6).
resynchronization, myocardial perfusion, drug delivery and
The remaining left ventricular veins are so variable
gene therapy. A good working knowledge of cardiac venous
that they are best described by location (see the next section).
anatomy is essential to all of these procedures (21,22).
Descriptive phrases such as “posterior vein” and “anterolateral
Unlike the CS, which is consistent in structure and posi-
vein draining to the AIV” are clinically useful and avoid con-
tion, the epicardial veins vary considerably in size, number and
fusion. These veins may drain into the CS and/or the GCV.
location. The surface veins of the heart resemble the superficial
There are several other named cardiac veins that may be
veins of the forearm: there are many individual variations
visualized during contrast venography. The small cardiac vein
and a few consistent features. While this has caused a lack of
lies in the groove between the RA and right ventricle, parallel to
consistency in nomenclature, uncertainty in the interpretation
the right coronary artery (RCA). It drains the posterior RA and
of venograms can be avoided by recognizing the few consistent
the right ventricle, entering the RA directly or joining the CS
features and then describing remaining veins according to their
near its ostium. Several anterior cardiac veins drain from the
location (Table 15.2).
front of the right ventricle directly into the RA.
The oblique vein of marshall (OVM) is also known as the
Recognizing Veins with Consistent Features oblique vein of the left atrium. It is a short, distinctive vessel
Two large veins are invariably present. They should be identi- that lies diagonally across the posterior wall of the LA. The
fied in every case because they provide reliable landmarks that OVM enters the CS immediately downstream from the valve of
orient the rest of the venous anatomy. The anterior interventric- Vieussens. It extends superiorly as the ligament of Marshall (a
ular vein (AIV) lies in the anterior interventricular groove, par- bundle of fibrous tissue, muscle fibers, nerve tissue and fat) to
allel to the left anterior descending (LAD) coronary artery. As it the left subclavian vein. The CS, the OVM and the ligament of
flows from the apex to the base of the heart, the AIV receives Marshall are all remnants or the left horn of the embryonic
branches from the anterolateral left ventricular free wall and the sinus venosus. Incomplete regression is the basis for persistent
interventricular septum. At the base of the heart, near the origin left superior vena cava (PLSVC).
of the LAD, the AIV turns sharply to the left and follows the
circumflex artery around the atrioventricular groove to the back Recognizing the Variable Veins
of the heart, where it empties into the CS. The segment in the Variable veins of the LV fall into three general groups. Posterior
atrioventricular groove is called the GCV. The GCV then usually veins drain to the CS just upstream from the MCV. Lateral veins
receives tributaries from the lateral LV free wall before empty- usually drain into the GCV. Lateral veins are located near the
ing into the CS. The GCV ends at the valve of Vieussens (where marginal branches of the circumflex artery. Anterolateral veins
the CS begins). The posterior interventricular vein (PIV) is the usually drain to the AIV; their location is similar to diagonal
venous equivalent of the posterior descending artery (PDA). It branches of the LAD.
[7/7/010/21:13:41] [157–172]
Figure 15.6 Identifying cardiac veins. (A) Typical venogram in the anteroposterior projection. (B) Interpretation should begin by identifying
the loop formed by consistent veins: the AIV, the PIV, and the GCV. These consistent veins plus the CS form a continuous loop that is easily
recognized. Abbreviations: AIV, anterior interventricular vein; PIV, posterior interventricular vein; GCV, great cardiac vein; CS, coronary
sinus.
Surface Collateral Patterns The PIV forms two of the basic collateral loops. The first,
The surface veins of the LV are connected by a network of the AIV-PIV loop, is almost always present. It is important
collateral channels, most of which are too small to be visualized because it establishes the position of the apex and the plane of
by venography (Fig. 15.7). Four of these channels make collat- the interventricular septum. Identifying the AIV-to-PIV loop
eral loops that are large enough and consistent enough to be prevents mistakes in localization. The second loop that includes
clinically useful. By understanding these loops, the angiogra- the PIV is the apical-lateral loop. Most patients have a large
pher can accurately localize any vein, regardless of imaging apical-lateral branch that connects the PIV to posterior and
projection. The collateral loops can also be exploited for guide- lateral veins. This loop is important because it helps to distin-
wire and lead placement. Loop 1 connects the AIV and PIV at guish apical from lateral locations. The apical-lateral loop also
the apex. Loop 2 connects the apical-lateral branch of the PIV to provides a “back door” to the lateral left ventricular free wall
posterior and lateral veins. Loop 3 connects posterior to lateral that can be utilized for cardiac resynchronization.
veins. Loop 4 connects lateral to anterolateral veins. The posterior-to-lateral loop often involves large
branches can be useful for left ventricular lead placement.
The anterolateral-to-lateral loop connects the AIV to lat-
eral veins. These branches are less desirable for cardiac
resynchronization because they tend to be small and located
near the phrenic nerve.
The ability to recognize four basic collateral loops is the
key to rapid identification of the epicardial veins and accurate,
three-dimensional localization of venous structures (Fig. 15.8).
The Pericardiophrenic Vein

Collateral filling of the left periocardiophrenic vein is some-
times seen during venography in patients with a history of
heart surgery. This is not a natural collateral loop, but the result
of postoperative inflammation, adhesions and neovasculariza-
tion. The pericardiophrenic vein is easily recognized by its
characteristic trajectory. It is a useful landmark because it
identifies the course of the phrenic nerve.
The phrenic nerve, along with the pericardiophrenic
artery and vein, form a bundle that courses over the lateral
wall of the LV to the left hemidiaphragm (Fig. 15.9). While
filling of the pericardiophrenic vein along the cardiac border is
Figure 15.7 Schematic diagram of the surface venous system. usually faint, the straight-line course is easily extrapolated. The
This illustration depicts numerous, interconnecting collateral path- pericardiophrenic vein can be catheterized intentionally to
ways. Source: From St. Jude Medical, Inc. mark the phrenic nerve during left ventricular lead placement
(23). It can also be catheterized unintentionally (24–26).
[7/7/010/21:13:41] [157–172]
Figure 15.8 Typical branches and collateral loops. By looking for collateral loops, what initially appears to be a nest of crisscrossing vessels
is easily interpretable. Epicardial veins of the left ventricle include anterior interventricular (a), posterior interventricular or middle cardiac (b),
posterior (c), lateral (d and e), anterolateral (f), and great cardiac (g). Collateral loops include anterior interventricular vein to PIV at the apex
(1), apical-lateral branch of PIV (2), posterior to lateral (3), and anterolateral to lateral (4). Abbreviation: PIV, posterior interventricular vein.
operators can struggle for hours. The difference lies not in

manual skill, but in a clear understanding of the anatomy.
Experienced operators make gentle and purposeful movements
toward a clearly defined mental target, utilizing known land-
marks as navigation aids. Novice operators tend to probe the
RA repeatedly in hopes that the catheter will eventually enter
the CS.
This section reviews a simplified approach right atrial
anatomy that allows rapid catheterization of the CS. We will
discuss general principles rather than specific catheters because
the choice of catheters is much less important than the knowl-
edge of where they need to go.
Two Key Steps

Successful catheterization of the CS is a two step process. First,
the operator must locate the ostium. This step requires specific
maneuvers to align the catheter with the axis of the CS ostium.
Figure 15.9 Dissection showing the course of the pericardio- Second, the operator must negotiate the body of the CS and
phrenic vein. The phrenic nerve, along with the pericardiophrenic enter the GCV. This step requires a different set of maneuvers
artery and vein, form a bundle that courses over the lateral wall of because the CS ostium is not aligned with the CS body or the
the left ventricle to the left hemidiaphragm. Source: From Bassett GCV.
Collection, Stanford University (http://ianatomy.stanford.edu). CS interventions, including cardiac resynchronization,
are almost always performed with subclavian or internal jug-
ular access. Therefore, we will discuss catheter manipulations
appropriate for an upper body approach (catheters entering via
The pericardiophrenic vein connects superiorly to the left the SVC). The manipulations are the same for left-sided and
subclavian vein and inferiorly to the left gastric vein or the right-sided approaches. The direction of rotation—clockwise or
inferior vena cava (IVC). Since it forms a collateral connection counterclockwise—is from the operator’s perspective, looking
between abdomen and thorax, it can become dilated when the down the shaft of the catheter.
superior vena cava (SVC) is occluded. Pacing leads and central
venous catheters have been inadvertently placed in dilated
pericardiophrenic veins, sometimes resulting in tamponade
(27–29). Finding the CS Ostium
Difficulties finding the CS usually arise from a lack of famil-
iarity with the orientation of the CS and great veins within the
TECHNIQUE OF CS CATHETERIZATION RA. Fortunately, these relationships are very consistent.
Experienced operators can catheterize the CS in a matter of The right atrial anatomy in the vicinity of the CS can be
seconds using almost any catheter, with few exceptions. Novice simplified. The SVC enters the posterior aspect of the RA and is
[7/7/010/21:13:41] [157–172]
oriented anteriorly, toward the tricuspid valve. The IVC, on the To enter the CS ostium, the first step is to advance the
other hand, is directed toward the fossa ovalis of the atrial catheter to the vicinity of the tricuspid valve. At the tricuspid
septum. The CS enters the RA between the IVC and the tricus- annulus, the catheter is rotated counterclockwise (from the per-
pid valve, with its opening is directed anteriorly toward the spective of the operator, looking down the shaft of the catheter)
tricuspid valve. These angles provide two distinct flow streams until the tip contacts the heart. The catheter is then slowly
during fetal life; oxygenated blood from the IVC flows prefer- withdrawn while maintaining gentle counterclockwise torque.
entially toward the LA, while deoxygenated blood from the The catheter tip will ride over the ridge of the tricuspid annulus
SVC and CS flow to the right ventricle. and drop into a shallow recess between the tricuspid annulus
Two valve-like structures help to maintain the divided and the Eustachian ridge. At that point, small injections of
flow streams within the RA. The Eustachian valve (and its contrast media should indicate the position of the CS. Forward
continuation as Chiari tissue) is present to varying degrees in advancement should engage the ostium.
the adult. It arises along the anterior margin of the IVC and Counterclockwise rotation is essential. Clockwise rotation
continues superiorly as the Eustachian ridge. The Eustachian causes the catheter tip to engage the Eustachian ridge and
valve and ridge help to funnel IVC flow toward the fossa valve, deflecting away from the CS and into the IVC. Even if
ovalis. The Thebesian valve covers the ostium of the CS to the catheter approaches the CS, clockwise rotations will cause
varying degrees. It always arises from the posterior margin of the tip to be deflected by the Thebesian valve, away from the
the CS ostium and directs flow toward the right ventricle. CS and into the right ventricle. The CS cannot be entered with
Between the IVC and the tricuspid annulus lies a recess clockwise rotation.
of right atrial free wall known as the Eustachian fossa. The If the CS is not entered immediately, test injections of
Eustachian fossa, the tricuspid annulus, and the Eustchian contrast will indicate one of two likely catheter locations. If the
ridge are easily identifiable landmarks that aid in localizing tip is in the Eustachian fossa, it is too inferior but a useful
the CS. landmark has been identified. The operator knows to repeat the
counterclockwise maneuver from a slightly more superior
starting point. If the catheter tip is in the body of the RA, it is
Catheter Manipulations to Engage the CS Ostium
too superior or too short to reach the tricuspid annulus. The
Catheters inserted via the SVC will not be aligned with the
operator knows to advance the catheter and repeat the coun-
ostium of the CS, so reorientation is always necessary
terclockwise maneuver from a more inferior starting point.
(Fig. 15.10). It is important to remember that the SVC does
Telescoping catheter systems are very helpful, especially
not enter the RA vertically, as it appears on two-dimensional
in patients with chamber dilatation. A typical telescoping sys-
fluoroscopy.
tem includes an outer guide catheter and an inner angiographic
catheter. Each catheter can be advanced and/or rotated inde-
pendently, with the combination of movements providing a
nearly complete range of tip trajectories. In most cases, the
outer guide catheter provides the anterior and superior starting
position for the angiographic catheter. The outer guide catheter
can be directed toward the anterior RA using clockwise rotation
while the inner angiographic catheter is directed toward the CS
using counterclockwise rotation. The addition of a guidewire
creates a triple telescoping system that is useful in cases of
extreme right atrial enlargement (Fig. 15.11).
CS catheterization is performed under fluoroscopic guid-
ance. The anteroposterior (AP) projection is usually sufficient
and has several advantages including lower x-ray dose rate,
more comfortable posture for the operator, and better preser-
vation of the sterile field. Angulated projections are not neces-
sary but sometimes useful. The left anterior oblique (LAO)
projection, preferred by some operators, provides left-right
perspective. The right anterior oblique (RAO) projection pro-
vides anterior-posterior perspective. The 308 RAO projection
often demonstrates a triangle of epicardial fat in the atrioven-
tricular groove at the base of the heart. This “fat pad sign”
provides a very reliable landmark in cases where the CS ostium
is not easily located. In difficult cases, the venous phase of
coronary arteriography can also help to localize the CS.
The femoral approach to CS catheterization is used for
diagnostic electrophysiology, blood sampling, and angiogra-
Figure 15.10 Correct catheter rotation. Counterclockwise rotation phy procedures that do not require the backup support needed
(from the operator’s perspective, looking down the catheter) allows to deliver pacing leads into branch veins. A variety of pre-
the catheter tip to enter the CS ostium coaxially and without formed and deflectable catheters is available for this purpose.
encountering the Thebesian valve. The Eustachian ridge and In addition, a number of conventional angiographic catheter
Thebesian valve guide the catheter tip toward the CS ostium. shapes can be utilized with or without a guidewire. The
Abbreviation: CS, coronary sinus.
femoral approach requires clockwise rotation to enter the CS
ostium (Fig. 15.12).
[7/7/010/21:13:41] [157–172]
Figure 15.11 Telescoping catheter technique for CS cannulation.

The outer delivery system is directed to the anterior right atrium, the
angiographic catheter is “aimed” toward the CS, and a steerable,
hydrophilic guidewire is advance into the CS. The angiographic
catheter can then be advance over the quidewire, followed by the
delivery system over the catheter. This is a very effective technique
in patients with marked right atrial enlargement. Abbreviation: CS,
coronary sinus. Figure 15.12 Coronary sinus catheterization from the femoral
approach. Catheter rotation to enter the ostium is clockwise (oppo-
site the superior approach). Source: From Ref. 30.
Entering the GCV occurs when the tip enters a valve cusp and retrograde force
After locating the CS ostium, the operator must negotiate closes the valve. The valve of Vieussens can produce so much
several changes in direction to enter the body of the CS and resistance to catheter advancement that it is misinterpreted as a
the GCV. Difficulty advancing catheters comes from using a fixed stenosis or an occlusion. The valve will not yield to force,
two-dimensional fluoroscopic image to navigate a three-dimen- but it can always be crossed by gently probing with a steerable,
sional structure (Fig. 15.13). This is resolved by understanding hydrophilic guidewire. The guidewire will hold the pliable
the anatomy. Forceful advancement of catheters can cause valve aside, allowing effortless passage of catheters and delivery
dissection or perforation. systems.
By venography in the LAO projection the CS appears to It is important to learn how to catheterize the GCV
form a flat, C-shaped loop around the mitral annulus. How- because most therapeutic procedures depend on a guide cath-
ever, from the RAO perspective, significant changes in direc- eter or delivery system that has been inserted deeply enough to
tion are apparent. To enter the CS, a catheter is directed down, be stable. The initial insertion should be beyond the area of
left and posterior using counterclockwise rotation (from a supe- interest. Subsequent manipulations are more precise because
rior approach). After crossing the contractile portion of the CS, pull-back removes slack and improves stability while avoiding
the catheter must be redirected upward and anterior using injury.
clockwise rotation. The catheter will eventually turn rightward
with the great cardiac, reaching the base of the heart at the
anterior interventricular groove. Complications of CS Catheterization
Resistance to catheter advancement at the valve of Vieus- Complications of CS catheterization include dissection, perfo-
sens is common, easily recognized, and easily overcome. Small ration and contrast extravasations. Compared with other cham-
injections of contrast provide important clues to location. The bers, the CS is relatively thin-walled easily traumatized (31,32).
valve of Vieussens is best recognized as an abrupt cutoff of The branch veins are even thinner and more fragile. Injuries
retrograde-injected contrast approximately 30 mm from the CS occur at predictable locations and are always avoidable.
ostium. If a OVM is visualized, the valve of Vieussens will be Dissections usually occur when a catheter or pacing lead
located immediately upstream. Finally when CS contractions are is forcefully advanced to overcome resistance at valves, branch
observed, the valve will mark the upstream extent of the con- points, or changes in direction (Fig. 15.14). Common locations
tractile portion of the CS. Resistance to catheter advancement include the origin of the MCV, the origin the OVM, the valve of
[7/7/010/21:13:41] [157–172]
Figure 15.13 Three-dimensional shape of the CS and GCV. During catheterization, the operator must navigate three-dimensional anatomy
with a two-dimensional fluoroscopic image. (A) Three-dimensional perspective. Volume-rendered reconstruction from multislice computed
tomography demonstrating the CS and GCV with several changes in direction. A slight annular constriction (small arrow) marks the junction of
the GCV and the CS. (B) Two-dimensional perspective. Fluoroscopy during the venous phase of coronary angiography. To enter the CS, a
catheter must be directed down, left and posterior using counterclockwise rotation (from a superior approach). After crossing the contractile
portion of the CS, the catheter must be redirected upward and anterior using clockwise rotation. The catheter will eventually turn rightward
with the great cardiac, reaching the base of the heart at the anterior interventricular groove. Abbreviations: GCV, great cardiac vein; CS,
coronary sinus. Source: From Ref. 35.
Figure 15.14 Common locations for CS injury. (A) Catheter advancement into the wall of the CS at the MCV. The axis of force (dashed
arrow) is not aligned with the CS. Injury can be avoided by using a curved catheter or steerable guidewire to negotiate the acute change in
direction needed to stay coaxial with the CS. (B) Localized dissection caused by catheter advancement into the wall of the CS at the MCV.
The axis of catheter advancement (arrow ) is nearly perpendicular to the wall of the CS. (C) Localized dissection at the origin of the oblique
vein of Marshall. (D) Localized dissection caused by forceful catheter advancement into the valve of Vieussens. The dissection extends into a
large lateral vein. (E) Catheter advancement into a small lateral vein. The axis of catheter force (dashed arrow) is directly into the branch vein,
not into the great cardiac vein. (F) Localized perforation caused by catheter advancement into a small lateral vein. Abbreviations: MCV,
middle cardiac vein; CS, coronary sinus.
[7/7/010/21:13:41] [157–172]
Vieussens, and at the origin of lateral branch veins. Dissections

can also result from balloon inflation in an undersized branch
vein. Perforations occur when a catheter inadvertently enters a
branch vein and force is directed perpendicular to the vessel
wall. Contrast extravasations can also occur during balloon-
occluded venography. The usual mechanism involves injection
through an end-hole catheter when the tip is against a vessel
wall.
Most catheter-induced CS injuries are clinically silent.
They are recognized during venography as contrast extrava-
sations that may appear menacing but are usually benign.
Contrast extravasations are usually contained by the epicardial
fat surrounding the CS. A persistent “stain” indicates that
contrast does not communicate with the pericardial space.
The operator should stop injecting at the first sign of contrast
extravasation, mainly because extravasations can obscure the
visualization of branch anatomy. Retrograde dissections close
spontaneously, with blood flow, when the offending catheter is
removed (33). The most significant consequence of dissection is Figure 15.15 Catheters for coronary sinus cannulation. A sample
that it can prevent cannulation of branch veins needed for of preformed and deflectable catheters and delivery sheaths from
optimal cardiac resynchronization. Perforations involving the one manufacturer. Source: From Medtronic, Inc.
low-pressure cardiac venous system usually seal spontaneously
when the catheter is removed. Tamponade is a rare complica-
tion of CS catheterization—rare enough that it should alert the
operator to an alternate explanation such as perforation of the venous anatomy. Special catheter shapes may help operators to
right ventricular apex by a pacing lead. Branch vein obstruc- overcome technical challenges in patients with advanced heart
tions are well tolerated because the cardiac venous drainage failure and chamber dilatation, but they are usually not unnec-
system is richly collateralized. essary. It is more important to remember that, even in heart
Injury to the CS is not necessarily a reason to terminate a failure, the anatomical relationships among structures remain
procedure. After recognizing an injury, the operator can safely constant. The CS is a large structure—approximately 10 mm
continue as long as the true lumen is identifiable. This may diameter at the ostium and larger in heart failure—that always
require a brief waiting period (5–10 minutes), a different choice of empties into the same location of the RA. In most cases, the CS
catheters and guidewires, and alternate imaging projections. can be cannulated easily with a few conventional catheters.
With extensive dissection, the operator may suspend the proce- No matter which catheters are selected, the objective is to
dure and return after allowing time (days to weeks) for healing. place a delivery system into a deep, stable, coaxial position
Most catheter-induced CS injuries heal without intervention. within the body of the CS. This provides the starting point for
Injury can be avoided by understanding the anatomy, by subsequent operations such as balloon-occluded venography,
utilizing small injections of contrast for orientation, and by selective catheterization of branch veins, and placement of left
choosing curved catheters or steerable guidewires to negotiate ventricular pacing leads.
changes in direction and remain coaxial with the vessel lumen.
BALLOON OCCLUSION VENOGRAPHY
Rationale
Choice of Catheters
Balloon-occlusion venography is the best method for visualiz-
CS catheters have evolved mainly from a need to provide CRT.
ing the cardiac veins at the time of left ventricular lead place-
Left ventricular pacing leads are installed through guide cath-
ment. The objective is to identify the branch veins that are most
eters that provide the backup support needed to “wedge” leads
ideal for placement of a left ventricular pacing lead. Optimal
securely into branch veins. Manufacturers now offer complete
lead positions represent the best compromise of anatomically
delivery system packages that include a variety of preformed
desirable location, stimulation threshold, stability, deliverabil-
and deflectable guide catheters, shaped catheters for CS can-
ity, and absence of diaphragmatic stimulation. To achieve
nulation, catheters for subselective cannulation of branch veins,
optimal lead positions, the operator must visualize all potential
guidewires and other tools (Fig. 15.15).
target veins and establish priorities (best, second best, accept-
Delivery systems are usually advanced coaxially over
able, and unacceptable). Balloon-occluded contrast venography
smaller, steerable catheters that are used to access the CS.
is the only method of consistently visualizing all potential
Some operators prefer deflectable electrophysiology catheters,
target veins. It is the cornerstone for effective CRT.
while others prefer fixed-curve angiographic catheters. Angio-
graphic catheters offer several advantages including low cost
and availability in a variety of shapes. Angiographic catheters Catheter Technique
also permit injections of contrast and the use of guidewires. Adequate venography requires a coaxial catheter alignment to
These attributes make angiographic catheters ideal for negoti- avoid subintimal injection, a stable balloon position during
ating complex anatomy. contrast injection, and complete visualization of all branch
Hardware refinements have made cardiac resynchroni- veins. The suggested technique reliably achieves these goals.
zation safer and more reliable, but an extensive array of The first step in balloon-occlusion venography is to
catheters is no substitute for a good working knowledge of advance a guide catheter or delivery system beyond the valve
[7/7/010/21:13:41] [157–172]
Figure 15.16 Balloon occlusion venography. (A) Guide catheter (block arrow) is advanced over a soft angiographic catheter (arrow) to the
great cardiac vein. (B) Balloon catheter is inserted and then retracted to a position just upstream from the valve of Vieussens (dashed arrows).
All filling downstream from the balloon is via collateral channels.
of Vieussens and into the GCV. It is often desirable to perform convenient working view that minimized radiation exposure.
contrast injections with the balloon in the GCV because this The AP projection alone is often sufficient. The RAO view pro-
structure is noncontractile and small enough to be easily vides additional apex-to-base perspective. The LAO view
occluded. This is important because the GCV is straight, non- provides additional superior-inferior perspective, and is best
contractile, and small enough to permit complete occlusion by a for displaying the takeoff of lateral veins draining the lateral
balloon catheter. Balloons inflated in the CS tend to be unstable left ventricular free wall.
and are easily ejected into the RA during contrast injection. Different projections can be reconciled by identifying the
Retrograde injections into the CS can close valve of Vieussens, two planes: the plane of the interventricular septum outlined
creating back pressure that ejects the balloon into the RA. The by the anterior and PIV, and the plane of the mitral valve
balloon catheter is inserted to the tip of the delivery system, outlined by the CS and GCV. Three-dimensional perspective
then “unsheathed” by retracting the delivery system. Alterna- can be obtained by visualizing the LV as a spherical fishbowl,
tively, the balloon catheters may be inserted over a guidewire. with veins curving across the outer surface and the neck
Small injections of contrast should confirm that the balloon tip representing the mitral valve plane (Figs. 15.17 and 15.18).
is beyond the valve of Vieussens, coaxial with the lumen, not in
a branch vein, and not against a vessel wall. A good target for
balloon inflation is the GCV just above the valve of Vieussens. NONINVASIVE IMAGING OF THE CS
The mild annular narrowing often present at this location helps AND CARDIAC VEINS
to stabilize the balloon, prevent balloon migration or contrast One of the most challenging aspects of cardiac venous catheter-
reflux (Fig. 15.16). The balloon is then inflated until test ization is navigating a three-dimensional structure while
injections of contrast indicate complete occlusion. With the guided by two-dimensional fluoroscopic images. Noninvasive
balloon inflated, images are acquired as 10 to 20 mL of contrast imaging modalities compliment catheter-based angiography by
is manually injected. Complete visualization of the cardiac providing spatial orientation. They also show how the cardiac
venous system is facilitated by extensive vein-to-vein collater- venous system is related to other cardiac structures—including
als. The injection must be rapid enough to produce collateral structures that are not visible by fluoroscopy.
filling of all branch veins. A reliable indicator of adequate The CS is routinely visualized by transthoracic, trans-
filling is complete opacification of the MCV, which is invariably esophageal and intracardiac echocardiography. Unfortunately,
present and easily recognized by its location in the posterior these modalities provide two-dimensional images of limited
interventricular groove. Undiluted contrast produces superior portions of the CS, factors that limit their ability to support
image quality but in some cases a 60% dilution reduces viscos- therapeutic procedures. Real-time, three-dimensional echocar-
ity and allows more rapid injections needed to fill branch veins. diography may overcome some of these limitations (34). Intra-
The balloon may remain inflated long enough to obtain veno- vascular ultrasound (IVUS) studies of the coronary arteries may
grams in multiple projections and to study the images for demonstrate the major epicardial veins, which must be recog-
completeness. Prolonged balloon occlusion of the GCV is nized as normal structures.
remarkably well tolerated, thanks to a rich collateral network Multislice computed tomography has been utilized to
that includes the Thebesian venous system. depict the cardiac veins and define the range of variability
(9,35). While it does require iodinated contrast and radiation
exposure, multislice computed tomography can provide high-
Image Acquisition spatial-resolution images of the cardiac venous system quickly
Venography should be performed in as many projections as and reliably. Multislice computed tomography has become a
necessary to fully define the anatomy. The AP projection is a legitimate tool for noninvasive preprocedure planning (Fig. 15.19).
[7/7/010/21:13:41] [157–172]
Figure 15.17 Projections for venography. Typical venograms in 308 RAO, AP and 508 LAO projections (top row). The same images annotated
to indicate the anterior interventricular vein–posterior interventricular vein loop that establishes the plane of the interventricular septum (dark
gray dashed lines) (bottom row). The semicircle formed by coronary sinus and great cardiac vein establishes the plane of the mitral annulus
(white dashed lines). The RAO projection maximizes the base-to-apex dimension (white arrow) and provides base-to-apex perspective. The
AP projection is a convenient working view. The LAO projection provides superior-inferior perspective (curved white arrow) and is best for
depicting the takeoff of lateral branch veins. Abbreviations: AP, anteroposterior; RAO, right anterior oblique; LAO, left anterior oblique.
Figure 15.18 Visualizing epicardial veins. The sur-

face veins are distributed over the free hemisphere of
left ventricle. Abbreviations: RAO, right anterior obli-
que; AP, anteroposterior; LAO, left anterior oblique.
Cardiac magnetic resonance imaging is another noninva- ABNORMALITIES OF THE CS

sive modality that provides clear depictions of the cardiac Persistent left superior vena cava (PLSVC) is the most common
venous system and its anatomical variations without the need anomaly of the thoracic veins, occurring in 0.3% of general
for iodinated contrast or ionizing radiation. The use of cardiac population and in about 5% of patients with congenital heart
magnetic resonance imaging for preprocedure planning may disease (37–41). The hallmark finding is a markedly dilated CS
reduce fluoroscopy time and increase the likelihood of success due to increased flow. PLSVC is usually an incidental finding
(Fig. 15.20) (36). by echocardiography or chest CT, but it may be discovered
[7/7/010/21:13:41] [157–172]
Figure 15.19 Noninvasive evaluation of CS anatomy using 64-slice computed tomography. Volume-rendered reconstructions demonstrate
the cardiac venous system and its relation to the mitral annulus and the left circumflex coronary artery (Cx). Abbreviation: CS, coronary sinus.
atrial septal defect. Unroofed CS is commonly associated with a

persistent left SVC. When a left SVC is present, venous blood
enters the LA and produces cyanosis. Unroofed CS is also
associated with other forms of complex congenital heart dis-
ease, usually heterotaxy syndromes. The clinical manifestations
of unroofed CS include right heart failure from chronic volume
overload, cyanosis and paradoxical embolism. The diagnosis
can be established by transthoracic or transesophageal echo-
cardiography and catheter-based angiography, but MRI and CT
are superior imaging modalities. Catheter-based interventions
aimed at covering the defect or occluding the left vena cava
have been described (51–53).
CS diverticulum is another malformation of the CS that
can be associated with arrhythmias and sudden cardiac death
(54,55). A CS diverticulum is usually discovered by contrast
Figure 15.20 Visualization of the cardiac venous system using venography at the time of electrophysiology procedures for
cardiac magnetic resonance imaging. Three-dimensional recon-
posterior-septal accessory pathways (Fig. 15.21) (56). Arrhyth-
struction of the heart (left) and multiplanar reformatted images
(right). Abbreviations: LA, left atrium; RA, right atrium; RV, right mias are mediated by muscle bridges joining LV, CS wall, and
ventricle; LV, left ventricle; CS, coronary sinus; RCA, right coronary atria. These pathways can have short refractory periods that
artery; PIV, posterior interventricular vein. Source: From Ref. 36. allow rapid conduction of atrial fibrillation. Rarely, CS diver-
ticulum is an incidental finding (57).
while passing catheters or pacing leads from the left subclavian EMERGING CS INTERVENTIONS
or jugular approach (catheters reach the right heart without Mitral Annuloplasty
passing though the SVC). The unusual trajectory can make The proximity of the CS to the mitral valve annulus has invited
manipulation of pacing leads difficult. PLSVC can also be researchers to explore percutaneous mitral annuloplasty. The
discovered during right heart catheterization from the femoral goal is to avoid the risks of surgical annuloplasty in patients
approach. The markedly enlarged CS is easily entered, and with functional mitral regurgitation secondary to dilated car-
catheters may appear to enter the pulmonary artery; the true diomyopathy. CS annuloplasty mimics surgical annuloplasty
location is indicated by venous pressures and venous oxygen by shortening the anterior-posterior dimension of the mitral
saturations. Persistent left SVC can be thought of as a very large annulus. Several approaches have been tried.
OVM that has failed to regress. Blood flow with PLSVC is The VIKING and MONARC devices (Edwards Lifescien-
physiologic, so there are no clinical consequences. ces, Irvine, California, U.S.) consist of two nitinol anchor stents
Unroofed CS is a rare type of atrial septal defect in which connected by a spring-like bridge. One anchor is deployed in
the normal separation between the walls of the LA and the CS the GCV and the other near the CS ostium (58). The bridge
are missing (42–50). The CS forms a bridge between the left and spring contracts as biodegradable spacers slowly dissolve,
right atria, allowing the left-to-right shunt that is typical of an pulling the stents together and shortening the annulus. The
[7/7/010/21:13:41] [157–172]
Figure 15.21 CS diverticulum. Venogram in the LAO

view demonstrating a large diverticulum (D) with neck
entering the CS near the ostium (left). The anterior
outflow of contrast (arrow) suggests the presence of a
Thebesian valve (Theb). A PCV is also labeled. Veno-
gram in the LAO view demonstrating another large CS
diverticulum associated with Wolff-Parkinson-White syn-
drome and atrial fibrillation (right). Abbreviations: LAO,
left anterior oblique; CS, coronary sinus; PCV, posterior
cardiac vein. Source: From Refs. 55 and 56.
CARILLON device (Cardiac Dimensions, Kirkland, Wisconsin, The venous approach may facilitate percutaneous myocardial
U.S.) is similar, with two self-expanding anchors connected by gene transfer by achieving high local concentrations of vectors
a tensioning ribbon (59). The design of this device allows with minimal systemic exposure (64).
tension to be adjusted at the time of implantation.
Initial human trials with these devices have demon-
strated significant, short-term reductions in the amount of Cardiac Veins for Myocardial Perfusion
mitral regurgitation. Design enhancements for both devices The presence of arteriovenous communications within the
have improved mechanical stability and durability. The CS myocardium has inspired clinicians to utilize cardiac veins as
approach to mitral annuloplasty is certainly feasible, but two conduits to perfuse ischemic myocardium. In the 1940s Beck
issues remain. First, the CS frequently lies along the LA free arterialized the cardiac veins by constructing vein grafts from
wall well above the mitral annulus. It is not yet known if this the aorta to the CS and then ligating CS outflow. Unfortunately,
geometric variability affects the outcome of CS annuloplasty (it patients developed myocardial edema, hemorrhage and fibro-
may not). Second, the GCV frequently crosses the circumflex sis. The technique was eventually scaled back to localized
artery or a major marginal branch. Extrinsic compression by the arerialization of branch veins draining ischemic areas. Some
device can lead to ischemic events. patients with severe angina improved, but interest evaporated
Percutaneous mitral annuloplasty devices will undoubt- with the advent of coronary artery bypass surgery (65).
edly evolve. The long-term safety and efficacy of CS annulo- In the 1980s venous perfusion was briefly revisited.
plasty remains to be established in pivotal trials. Percutaneous, synchronized CS retroperfusion was developed
to manage unstable angina, and abrupt occlusion during cor-
onary angioplasty (66–70). The objective was to inject arterial
CS Reducer Stent blood into the CS during diastole, and allow venous drainage
A CS reducer stent has been developed as a treatment for during systole. CS retroperfusion failed because of cumber-
refractory angina in patients who are not candidates for con- some equipment, inexperience with CS catheterization, delays
ventional revascularization procedures (60). This device is initiating therapy, and limited benefit. Its main purpose—
based on the hypothesis that an increase in CS back pressure support during high-risk angioplasty—was eliminated by
can redistribute blood from nonischemic to ischemic myocar- stents.
dium via Thebesian channels. A feasibility study has demon- Percutaneous in situ coronary venous arterialization
strated significant reductions in angina and improvements in (PICVA) has been performed in a few patients with ungraftable
objective measures of myocardial perfusion, without adverse LAD disease (71). This procedure involves IVUS-guided punc-
events. While further study will be needed to validate the initial ture from LAD to AIV, placement of a stent to connect artery
findings and determine the mechanism of action, the CS and vein, and occlusion of venous outflow. The technical
reducer stent could become another catheter-based option for challenges and complications have been substantial (72,73).
patients with refractory angina. Nevertheless, PICVA demonstrates a creative approach to
venous perfusion.
A percutaneous, stent-based, LV-to-vein bypass proce-
Drug Delivery and Gene Therapy dure has been performed in animals (74). The procedure
The cardiac venous system provides an alternate route to appears to be technically feasible and capable of establishing
ischemic myocardium (61,62). Venous infusion can deliver both systolic inflow and diastolic outflow. Long-term conse-
high drug concentrations to targeted areas, making it ideal quences are not yet known.
for interventions to reduce ischemic and reperfusion injury. Patients with refractory angina who are not candidates
The cardiac veins could also be a route for delivering angio- for conventional revascularization will continue to stimulate
genic growth factors and stem cells to the myocardium (63). interest in the use of cardiac veins for myocardial perfusion.
[7/7/010/21:13:41] [157–172]
Unfortunately, there are serious concerns about the durability, 15. Eckardt L. Automaticity in the Coronary Sinus. J Cardiovasc
safety and efficacy of arterialized cardiac veins. The future of Electrophysiol 2002; 13:288–289.
venous arterialization is still in doubt. 16. Pejkovic B, Bogdanovic D. The great cardiac vein. Surg Radiol
Anat 1992; 14:23–28.
17. de Oliveira IM, Scanavacca MI, Correia AT, et al. Anatomic rela-
Contrast Recapture tions of the Marshall vein: importance for catheterization of the
It has been shown that the majority of the contrast injected into coronary sinus in ablation procedures. Europace 2007; 9:915–919.
a coronary artery can be captured from the CS before it enters 18. Piffer CR, Piffer MI, Zorzetto NL. Anatomic data of the human
the systemic circulation, potentially reducing the risk of con- coronary sinus. Anat Anz 1990; 170:21–29.
19. Karaca M, Bilge O, Dinckal MH, et al. The anatomic barriers in the
trast nephropathy (75).
coronary sinus: implications for clinical procedures. J Interv
Cardiol Electrophysiol 2005; 14:89–94.
Cardiac Endoscopy 20. Hill AJ, Coles JA Jr, Sigg DC, et al. Images of the human coronary
sinus ostium obtained from isolated working hearts. Ann Thorac
Infrared endoscopy is a novel (albeit invasive) strategy for
Surg 2003; 76:2108.
visualizing the CS and its valves (76).
21. Loukas M, Bilinsky S, Bilinsky E, et al. Cardiac veins: a review of
the literature. Clin Anat 2009; 22:129–145.
22. von Lüdinghausen M. Clinical anatomy of cardiac veins. Surg
CONCLUSIONS Radiol Anat 1987; 9:159–168.
The cardiac venous system has become a vital pathway to the 23. Vaseghi M, Cesario DA, Ji S, et al. Beyond coronary sinus
left heart. A good understanding of its anatomy, physiology, angiography: the value of coronary arteriography and identifica-
and catheterizations techniques is essential to CRT and invasive tion of the pericardiophrenic vein during left ventricular lead
arrhythmia management. Emerging therapies that depend on placement. Pacing Clin Electrophysiol 2005; 28:185–190.
cardiac venous catheterization include percutaneous mitral 24. Kogan AD, Malek M. Blind cannulation of the left pericardio-
annuloplasty, gene delivery, stem cell delivery, and retrograde phrenic vein: an unusual cause of diaphragmatic pacing. Pacing
myocardial perfusion. New interdisciplinary relationships will Clin Electrophysiol 1993; 16:356–359.
require the interventional cardiologist to be skilled at catheter- 25. McLellan BA, Jerman MR, French WJ, et al. Inadvertent Swan-
Ganz catheter placement in the left pericardiophrenic vein. Cathet
ization of the cardiac venous. The CS can no longer be ignored.
26. Wechsler RJ, Byrne KJ, Steiner RM. The misplaced thoracic venous
catheter: detailed anatomical consideration. Crit Rev Diagn Imag-
REFERENCES ing 1984; 21:289–305.
1. McAlister FA, Ezekowitz J, Hooton N, et al. Cardiac resynchroni- 27. Krishnan A, Cacciarelli A, Gibson D. Unusual complication of
zation therapy for patients with left ventricular systolic dysfunc- peripherally inserted central venous catheter placement: the left
tion: a systematic review. JAMA 2007; 297:2502–2514. pericardiophrenic vein. Pediatr Radiol 2004; 34:180–181.
2. Ansari A. Anatomy and clinical significance of ventricular The- 28. Colomina MJ, Godet C, Pellisé F, et al. Cardiac tamponade
besian veins. Clin Anat 2001; 14:102–110. associated with a peripheral vein central venous catheter. Paediatr
3. Ravin MB, Epstein RM, Malm JR. Contribution of thebesian veins Anaesth 2005; 15:988–992.
to the physiologic shunt in anesthetized man. J Appl Physiol 1965; 29. van Haeften TW, van Pampus EC, Boot H, et al. Cardiac tampo-
20:1148–1152. nade from misplaced central venous line in pericardiophrenic
4. Faxon DP, Jacobs AK, Kellett MA, et al. Coronary sinus occlusion vein. Arch Intern Med 1988; 148:1649–1650.
pressure and its relation to intracardiac pressure. Am J Cardiol 30. Winters KJ, Lasala JM, Eisenberg PR, et al. Modified heparin-
1985; 56:457–460. bonded catheter for cannulation of the coronary sinus from the
5. Singhal S, Khoury S. Images in clinical medicine. Imaging of femoral vein. Cathet Cardiovasc Diagn 1996; 39:433–437.
Thebesian venous system. N Engl J Med 2008; 359(7):e8. 31. Johnson WB, Mayotte M, Bailin S, et al. Incidence of coronary
6. Silver MA, Rowley NE. The functional anatomy of the human sinus dissection and perforation complications from coronary
coronary sinus. Am Heart J 1998; 115:1080–1084. sinus venograms in a large multicenter trial. Pacing Clin Electro-
7. Ortale JR, Gabriel EA, Iost C, et al. The anatomy of the coronary physiol 2003; 26:S56.
sinus and its tributaries. Surg Radiol Anat 2001; 23:15–21. 32. Walker S, Levy T, Paul VE. Dissection of the coronary sinus
8. Ho SY, Sánchez-Quintana D, Becker AE. A review of the coronary secondary to pacemaker lead manipulation. Pacing Clin Electro-
venous system: a road less travelled. Heart Rhythm 2004; 1: physiol 2000; 23:541–543.
107–112. 33. de Cock CC, van Campen CM, Visser CA. Major dissection of the
9. Jongbloed MR, Lamb HJ, Bax JJ, et al. Noninvasive visualization of coronary sinus and its tributaries during lead implantation for
the cardiac venous system using multislice computed tomogra- biventricular stimulation: angiographic follow-up. Europace 2004;
phy. J Am Coll Cardiol 2005; 45:749–753. 6:43–47.
10. Barcelo A, De La Fuente L, Stertzer S. Anatomic and histologic 34. Conca C, Faletra F, Chioncel O, et al. Coronary sinus visualization
review of the coronary sinus. Int J Morphol 2004; 22(4):331–338. by 3-dimensional real-time echocardiography. J Am Soc Echocar-
11. Chauvin M, Shah DC, Haissaguerre M, et al. The anatomic basis of diogr 2008; 21:371–376.
connections between the coronary sinus musculature and the left 35. Tops LF, Van de Veire NR, Schuijf JD, et al. Noninvasive evalu-
atrium in humans. Circulation 2000; 101(6):647–652. ation of coronary sinus anatomy and its relation to the mitral valve
12. D’Cruz IA, Johns C, Shala MB. Dynamic cyclic changes in coro- annulus: implications for percutaneous mitral annuloplasty. Cir-
nary sinus caliber in patients with and without congestive heart culation 2007; 115:1426–1432.
failure. Am J Cardiol 1999; 83:275–277. 36. Chiribiri A, Kelle S, Götze S, et al. Visualization of the cardiac
13. D’Cruz IA, Shirwany A. Update on echocardiography of coronary venous system using cardiac magnetic resonance. Am J Cardiol
sinus anatomy and physiology. Echocardiography 2003; 20:87–95. 2008; 101:407–412.
14. Antz M, Otomo K, Arruda M, et al. Electrical conduction between 37. Tahir T, Crouch E, Drake GB. Persistent left superior vena cava:
the right atrium and the left atrium via the musculature of the incidence, significance and clinical correlates. Int J Cardiol 2002;
coronary sinus. Circulation 1998; 98:1790–1795. 82:91–93.
[7/7/010/21:13:41] [157–172]
38. Alam M, Thaver GH, Misra A, et al. Persistent left superior vena 59. Schofer J, Siminiak T, Haude M, et al. Percutaneous mitral
cava: an incidental finding. J Am Coll Cardiol 2009; 53:1159. annuloplasty for functional mitral regurgitation: results of the
39. Ho KL, Tan JL. Persistent left sided superior vena cava. Heart CARILLON Mitral Annuloplasty Device European Union Study.
2006; 92:1046. Circulation 2009; 120:326–333.
40. Schulz HU, Mantke R, Freitag J, et al. Persistent left superior vena 60. Banai S, Ben Muvhar S, Parikh KH, et al. Coronary sinus reducer
cava. Lancet 2003; 361:560. stent for the treatment of chronic refractory angina pectoris: a
41. Gerber TC, Kuzo RS. Images in cardiovascular medicine. Persis- prospective, open-label, multicenter, safety feasibility first-in-man
tent left superior vena cava demonstrated with multislice spiral study. J Am Coll Cardiol 2007; 49:1783–1789.
computed tomography. Circulation 2002; 105(14):e79. 61. Jain AK, Smith EJ, Rothman MT. The coronary venous system: an
42. Nakatani S, Katsuki K, Miyatake K. Images in cardiology. alternative route of access to the myocardium. J Invasive Cardiol
Unroofed coronary sinus. Heart 2002; 87:278. 2006; 18:563–568.
43. Sueyoshi E, Sakamoto I, Mizowaki T, et al. Persistent left superior 62. Kar S, Nordlander R. Coronary veins: an alternate route to
vena cava into left atrium. Lancet 2006; 368:1270. ischemic myocardium. Heart Lung 1992; 21:148–157.
44. Piacentini G, Saffirio C, Marino B. Persistent left superior vena 63. von Degenfeld G, Raake P, Kupatt C, et al. Selective pressure-
cava into unroofed coronary sinus. Lancet 2006; 368:1963–1964. regulated retroinfusion of fibroblast growth factor-2 into the cor-
45. Kwok OH, Chan JK. Unroofed coronary sinus defect. Hong Kong onary vein enhances regional myocardial blood flow and function
Med J 2008; 14:331–332. in pigs with chronic myocardial ischemia. J Am Coll Cardiol 2003;
46. Huang XS. Images in cardiovascular medicine. Partially unroofed 42:1120–1128.
coronary sinus. Circulation 2007; 116:e373. 64. Hou D, Maclaughlin F, Thiesse M, et al. Widespread regional
47. Acar P, Arran S, Paranon S. Unroofed coronary sinus with per- myocardial transfection by plasmid encoding Del-1 following
sistent left superior vena cava assessed by 3D echocardiography. retrograde coronary venous delivery. Catheter Cardiovasc Interv
Echocardiography 2008; 25:666–667. 2003; 58:207–211.
48. Thangaroopan M, Truong QA, Kalra MK, et al. Images in cardio- 65. Beck CS, Stanton E, et al. Revascularization of heart by graft of
vascular medicine. Rare case of an unroofed coronary sinus: systemic artery into coronary sinus. J Am Med Assoc 1948; 137
diagnosis by multidetector computed tomography. Circulation (5):436–442.
2009; 119:e518–e520. 66. Hajduczki I, Kar S, Areeda J, et al. Reversal of chronic regional
49. Oyama N, Ooka T, Sasaki T, et al. Volume-rendering and endomyocardial dysfunction (hibernating myocardium) by synchron-
cardial views of partially unroofed coronary sinus with 64-slice ized diastolic coronary venous retroperfusion during coronary
multidetector CT. J Cardiovasc Comput Tomogr 2009; 3(5): angioplasty. J Am Coll Cardiol 1990; 15:238–242.
346–347. 67. Kar S, Drury JK, Hajduczki I, et al. Synchronized coronary venous
50. Low SC, Oliveira GR, Maki JH. Magnetic resonance imaging of retroperfusion for support and salvage of ischemic myocardium
unroofed coronary sinus. Heart 2009; 95(9):720. during elective and failed angioplasty. J Am Coll Cardiol 1991;
51. Heng JT, De Giovanni JV. Occlusion of persistent left superior 18:271–282.
vena cava to unroofed coronary sinus using vena cava filter and 68. Kar S, Barnett JC, Freedman RJ Jr., et al. Synchronized coronary
coils. Heart 1997; 77:579–580. venous retroperfusion. J Am Coll Cardiol 1994; 24:579–581.
52. Geggel RL, Perry SB, Blume ED, et al. Left superior vena cava 69. Kassab GS, Navia JA, March K, et al. Coronary venous retroper-
connection to unroofed coronary sinus associated with positional fusion: an old concept, a new approach. J Appl Physiol 2008;
cyanosis: successful transcatheter treatment using Gianturco- 104:1266–1272.
Grifka vascular occlusion device. Catheter Cardiovasc Interv 70. Gerber TC, Kantor B, Keelan PC, et al. The coronary venous
1999; 48:369–373. system: an alternate portal to the myocardium for diagnostic
53. Torres A, Gersony WM, Hellenbrand W. Closure of unroofed and therapeutic procedures in invasive cardiology. Curr Interv
coronary sinus with a covered stent in a symptomatic infant. Cardiol Rep 2000; 2:27–37.
Catheter Cardiovasc Interv 2007; 70:745–748. 71. Oesterle SN, Reifart N, Hauptmann E, et al. Percutaneous in situ
54. Guiraudon GM, Guiraudon CM, Klein GJ, et al. The coronary coronary venous arterialization. Report of the first human cathe-
sinus diverticulum: a pathologic entity associated with the Wolff- ter-based coronary artery bypass. Circulation 2001; 103:2539–2543.
Parkinson-White syndrome. Am J Cardiol 1988; 62:733–735. 72. Oesterle SN, Reifart N, Hayase M, et al. Catheter-based coronary
55. Blank R, Dieterle T, Osswald S, et al. Images in cardiovascular bypass: a development update. Catheter Cardiovasc Interv 2003;
medicine. Wolff-Parkinson-White syndrome and atrial fibrillation 58:212–218.
in a patient with a coronary sinus diverticulum. Circulation 2007; 73. Chowdhry MF, Davies J, McCance A, et al. Lack of durability of
115:e469–e471. surgical arterialization of coronary veins for the treatment of
56. Sun Y, Arruda M, Otomo K, et al. Coronary sinus-ventricular ischemic heart disease. J Card Surg 2005; 20:326–328.
accessory connections producing posteroseptal and left posterior 74. Raake P, Hinkel R, Kupatt C, et al. Percutaneous approach to a
accessory pathways: incidence and electrophysiological identifi- stent-based ventricle to coronary vein bypass (venous VPASS):
cation. Circulation 2002; 106(11):1362–1367. comparison to catheter-based selective pressure-regulated retro-
57. Thal S, Thai H, Juneman E, et al. Coronary sinus diverticulum infusion of the coronary vein. Eur Heart J 2005; 26:1228–1234.
complicating CRT device implantation. Pacing Clin Electrophysiol 75. Meyer M, Dauerman HL, Bell SP, et al. Coronary venous capture of
2008; 31:1184–1185. contrast during angiography. J Interv Cardiol 2006; 19(5):401–404.
58. Webb JG, Harnek J, Munt BI, et al. Percutaneous transvenous 76. Nazarian S, Knight BP, Dickfeld TL, et al. Direct visualization of
mitral annuloplasty: initial human experience with device implan- coronary sinus ostium and branches with a flexible steerable
tation in the coronary sinus. Circulation 2006; 113:851–855. fiberoptic infrared endoscope. Heart Rhythm 2005; 2:844–848.
[2/7/010/8:36:2] [173–193]
16
Coronary arterial anatomy: normal, variants,

and well-described collaterals
John P. Erwin and Gregory J. Dehmer
INTRODUCTION endothelium are dependent on vessel size. There are two

As with all human science, a strong understanding of normal major mechanisms, which regulate the endothelial barrier
physiologic function and many disease states requires a solid function; one involves cell-to-cell contacts allowing transendo-
knowledge base of the normal anatomy. During the perfor- thelial transport in the junctions between cells, the other
mance of coronary angiography, the anatomy of the epicardial involves transendothelial transport via vesicular transport
coronary arteries including collateral vessels is the main focus. directly through the cell.
However, with the development of percutaneous techniques to The next layer of an artery is the tunica media or simply
treat epicardial coronary artery stenoses, an understanding of media. It surrounds the internal elastic lamina and its compo-
the normal anatomic structure of a coronary artery is also sition depends on the type of artery. Large arteries have addi-
critical. Although this chapter will focus mainly on the anatomy tional circumferential layers of elastic tissue within the media
of the epicardial coronary system, a brief review of the struc- and are thus referred to as elastic arteries. The epicardial
ture of a normal coronary artery is relevant. coronary arteries are elastic arteries as are the carotids, cerebral
arteries and the aorta. However, at the point the epicardial
arteries turn into the myocardium, usually at a right angle from
CORONARY ARTERY ANATOMY the parent vessel, they become more muscular arteries with few
A normal coronary artery consists of three histologically dis- if any elastic fibers. In normal arteries, the vessel lumen diam-
tinct layers (Fig. 16.1). The innermost layer, which surrounds eter can be altered by contraction or relaxation of the medial
the lumen, is the tunica intima or simply the intima. It is vascular smooth muscles in response to a variety of systemic
composed of a single layer of endothelial cells in close prox- signals and locally released factors. However in certain disease
imity to the internal elastic lamina. The normal endothelial states, such as restenosis following balloon angioplasty, growth
layer has a central role in maintaining vascular health and has and hypertrophy with migration into the intima are the usual
three distinctive roles. First, it is a metabolically active secretory responses (8).
tissue. Endothelial cells secrete both vasodilator substances Finally, the media of all arteries is contained within a
such as prostacyclin, nitric oxide and endothelial-derived connective tissue layer called the tunica adventitia, or simply
hyperpolarizing factors and vasoconstrictor substances (endo- adventitia. In elastic arteries, this is demarcated by a layer of
thelin and vasoconstrictor prostanoids) (1–3). Endothelial cells elastic fibers termed the external elastic lamina. The adventitia
also produce and secrete von Willebrand factor, factor VIII contains a network of small blood vessels (vaso vasorum),
antigen, tissue factor, and tissue plasminogen activator, which which is responsible for the nutrition of the outer two-thirds
have important roles in coagulation and fibrinolysis. Moreover, of the artery. The inner third of the artery derives nutrition by
endothelial cells secrete certain structural components of the diffusion through the endothelium. The adventitia also contains
extracellular matrix such as elastin, glycosaminoglycans and nerves, which control the tone of the artery. In various locations
fibronectin and, along with smooth muscle cells, matrix metal- within the adventitia and associated with the outermost elastic
loproteinases, which are critical in arterial remodeling (4,5). layer are pressure receptors. These pressure or baroreceptors
Finally, certain growth factors that that control smooth muscle have a phasic discharge rate in harmony with the arterial
proliferation are secreted by endothelial cells (6). Second, the stretching associated with the pulse wave. Impulses of the
endothelium has an anticoagulant and antithrombotic surface. baroreceptors are integrated centrally and when increased or
Normal quiescent endothelial cells have an antithrombotic decreased cause appropriate alterations in many vascular beds.
surface that inhibits platelet adhesion and coagulation, but
when stimulated by cytokines or other inflammatory mediators
are capable of prothrombotic factors. Substances secreted by the BASIC ANATOMY OF THE CORONARY
endothelium include prostacyclin and nitric oxide, which affect CIRCULATION
platelet aggregation, include, plus antithrombin III, heparin-like The anatomy of the coronary arteries was described by the
molecules and tissue plasminogen activator (7). Although the French anatomist Raymond Vieussens almost 300 years ago.
endothelium normally exists in a functional state of balance Careful postmortem studies using injection techniques have
between thrombotic and antithrombotic factors, injury or provided greater detail and a more definitive description of the
inflammation enhances the prothrombotic state of the endothe- normal coronary anatomy. The Latin term “corona,” or crown,
lium. Finally, the normal endothelium provides a barrier to the aptly describes coronary arteries as the branches of these two
indiscriminate passage of blood constituents into the arterial arteries traverse the atrioventricular and interventricular sulci
wall. Fluid and macromolecular transport functions of the in the shape of a crown (9). In humans as well as birds, reptiles
[2/7/010/8:36:2] [173–193]
a portion of the obtuse margin of the heart between the LAD and
LCx. In one series, a ramus intermedius occurred in 37% of the
general population, and was considered a normal variant (14). In
another series of 150 hearts, Baptista et al. showed that the left
main bifurcated in 55%, trifurcated in 39%, and had a quad-
rification (ramus and separate diagonal) in 7% (15). A trifurca-
tion pattern was found in 60% of nonwhite females. The length
of the ramus varies from 20 to 50 mm and its relative length
varied from 21% to 50% of the length of the left ventricle (LV).
The normal location of the RCA ostium is more variable. It
usually arises from the middle of the right coronary sinus just
below the sinotubular junction of the right sinus of Valsalva.
However, it can arise from low near the valve to high near the
sinotubular ridge. The right and left coronary arteries are usu-
ally the only vessels arising immediately above the free margin
of aortic valve from the ascending aorta. However in up to 30%
of angiograms, the artery to the pulmonary outflow tract (conus
artery) originates as a separate ostium rather than its usual
position as a branch of the proximal RCA. Common variations
Figure 16.1 Normal human coronary artery (hematoxylin and of the location of the coronary ostia exist, which can have clinical
eosin stain at 40 power magnification) showing the intima (I), implications as will be discussed later in this chapter.
media (M), and adventitia (A). Source: Courtesy of Dr V.O.
Speights, Department of Pathology, Scott & White Healthcare.
Coronary Dominance
The term coronary dominance was introduced by Schlesinger in
1940 (16). The dominant coronary artery is the one from which
and mammals, the major epicardial vessels of the coronary the posterior descending coronary artery arises (Fig. 16.2). The
posterior descending artery (PDA) traverses the posterior inter-
circulation are the left main coronary artery and the right main
ventricular sulcus and supplies the posterior part of the ven-
coronary artery (RCA). The left and right coronary arteries
originate at the base (root) of the aorta from openings called the tricular septum and often a portion of the posterolateral wall of
the LV. There is variability in the literature regarding the
coronary ostia, which are located in the left and right sinus of
frequency of right dominance with sources varying from a
Valsalva, respectively. The ostia usually originate at the center
of each sinus just below or no more than 1 cm above the low of 70% to a high of 90% (18). In a right dominant circulation,
superior edge of the aortic cusp. The left coronary orifice the RCA crosses the interventricular groove and continues in the
atrioventricular groove beyond the origin of the PDA to supply
normally arises from the left sinus of Valsalva midway between
one or more posterolateral left ventricular branches. The artery
the posterior portion of the pulmonary artery and the left atrial
appendage just above the level of the free margin of the aortic to the atrioventricular node usually arises from the RCA at an
area called the crux, which represents the intersection of the
valve leaflet and generally below the sinotubular junction. The
interventricular and atrioventricular grooves on the inferior
left coronary ostium is usually single, giving rise to a short,
common left coronary artery (LCA) trunk (the left main artery) surface of the heart. This is noted on the posterior surface of
the heart by a small indentation or dimple.
that courses in the epicardial fat for distances varying from a
If the PDA arises from the terminal portion of the LCx,
few millimeters to several centimeters before bifurcating into
the left anterior descending (LAD) and left circumflex (LCx) the term left dominance is applied to the circulation. The
coronary arteries. The length of the left main artery as derived frequency of a left dominant circulation varies from 8% to
15% of individuals (18). In the remaining individuals, the
from pathologic examinations is 1.0 0.3 cm (10). Some
posterior septum is supplied by branches arising from both
premortem and postmortem angiographic studies have sug-
gested that patients with bicuspid aortic valves have shorter left the RCA and LCx. In this situation the circulation is said to be
“balanced” or codominant with dual posterior descending
main segments, but this is not universally accepted (11). No
arteries (one from the RCA and the other from the LCx) or no
correlation of left main coronary artery length with age, gender,
heart weight, extent of coronary artery disease or left ventric- clear PDA with multiple smaller branches arising from both
arteries. It is important to note that anatomic dominance does
ular wall thickness was found in one autopsy series (12).
not imply the vessel is of greater physiologic importance and
Histologically, the left main ostium lacks adventitia and has a
greater portion of elastic tissue than other segments of the thus is somewhat a misnomer. Although the RCA is most
frequently the dominant artery, the left coronary artery almost
coronary tree. This may account for some of the differences in
always supplies a greater myocardial mass (19).
response to coronary interventions involving this segment of
the left main. Moreover, since the left main ostium technically
lies within the wall of the aorta, it is subject to diseases affecting
the aortic wall such as syphilitic aortitis, radiation-induced LAD Artery
aortitis and Takayasu’s arteritis (13). The LAD is a direct continuation of the left main artery, with its
In some individuals, there is a trifurcation of the left main course along the anterior interventricular groove. When the
with a third branch arising in the crotch between the LAD and heart is viewed frontally, the LAD is seen as it curves around
LCx. This third artery, called a ramus intermedius or simply a and emerges from behind the pulmonary artery (Fig. 16.3). The
ramus branch acts functionally as a circumflex branch, supplying LAD in combination with the left main forms a curve that
[2/7/010/8:36:2] [173–193]
CORONARY ARTERIAL ANATOMY 175
Figure 16.2 Right and left dominant coronary anatomy.

The dominant coronary artery is the one from which the
posterior descending coronary artery arises. Abbrevia-
tions: RCA, right coronary artery; LAD, left anterior
descending; CX, circumflex artery; PDA, posterior
descending artery. Source: From Ref. 17, Mayo Foun-
dation for Medical Education and Research.
resembles a reverse “S” shape. The tight upper curve brings the (19). In contrast, the LAD may course well around the apex
LAD around the pulmonary artery to reach the uppermost (“wraparound” LAD) and supply a substantial portion of the
portions of the interventricular septum, and the lower portion posterior septum and even replace the PDA (20,21). Some
curves in the opposite direction as it follows the interventricular individuals may have dual LAD systems with one trunk almost
septum toward and usually around the apex. Several normal exclusively supplying the septum and the other trunk running
variations in the length and distribution of the LAD have been almost parallel giving rise to all of the diagonal arteries (22).
recognized. It is not essential for the LAD to reach the cardiac Along its course toward the cardiac apex, the LAD gives rise to
apex or to have well-defined branches to qualify as the LAD anterior septal perforating branches and diagonal branches. As
these branches arise from the LAD, the lumen diameter of the
LAD is progressively reduced toward the cardiac apex. Small,
branches may arise from the LAD to supply a small portion of
the anterior wall of the right ventricle with isolated reports of
larger vessels termed a right ventricular descending branch (23).
Diagonal Arteries
Diagonal artery branches arise from the LAD and course at
downward angles to supply the anterolateral free wall of the LV
(Fig. 16.3). Most individuals have one to three diagonal arteries,
but up to six small diagonals have been seen. The larger diag-
onal arteries arise from the upper portion of the LAD and the
caliber of the branches becomes progressively smaller as the
LAD approaches the apex. The diagonals roughly run parallel to
each other and also parallel to a ramus branch if one is present.
Septal Perforating Branches

In over 99% of individuals, the blood supply to the anterior
interventricular septum is from the LAD (24). These septal
perforating branches arise at nearly right angles from the LAD
and penetrate deep into the interventricular septum. In the
majority of individuals, there is no dominant septal artery, but
several proximal septal vessels of equal size. However, in 38% of
individuals, a large dominant septal perforator is present and is
Figure 16.3 3-D reconstruction of a 64-slice CT angiogram illus- usually, but not always the first septal (24). Septal perforator
trating the coronary tree in a left anterior oblique view with cranial arteries may bifurcate and trifurcate, but the branching pattern is
angulation. Aorta is marked by “Ao.” Abbreviations: MPA, main somewhat unordered and may take the appearance of a pitch-
pulmonary artery; LAD, left anterior descending artery; D1, first fork with multiple branches off one central point of the trunk.
diagonal artery; RCA, right coronary artery. Source: Courtesy of Septal arteries are occasionally seen arising from other arteries
Dr John Rumberger, Princeton Longevity Center. such as the first diagonal, proximal RCA or LCx and as
a separate ostium from the right sinus of Valsalva (25,26).
[Shaji][D:/informa_Publishing/Mukherjee_2400046/z_production/z_3B2_3D_files/978-1-8418-4664-
4_CH0016_O.3d] [16/7/010/16:28:31] [173–193]
LCx Artery
The LCx coronary artery arises from the left main artery at its
bifurcation and courses posteriorly under the left atrial append-
age to reach the left atrioventricular groove (Fig. 16.4). The
origin of the LCx is often at nearly a right angle to the left main,
but in some patients may have a greater or lesser degree of
angulation from the left main at its origin. As it continues, it
remains in the atrioventricular groove circumscribing the
mitral valve annulus giving rise to as many as four obtuse
marginal arteries. The extent and distribution of the LCx and
marginal arteries is usually reciprocal with that of the RCA. If
the LCx is has an extensive distribution over the posterior and
inferior walls, the RCA will usually be small with fewer
branches in this region and vice-versa. Atrial branches may
arise from the LCx coronary artery and supply the sinus node
in 37% of patients (27). Although the circumflex sinus node
artery usually arises near the origin of the LCx, in approxi-
mately 20% of individuals who have a circumflex origin of this
artery, the artery is an S-shaped vessel that originates from the
posterolateral branch of the circumflex.
Obtuse Marginal Arteries

The marginal arteries run parallel to the diagonal arteries and
ramus if one is present. The nomenclature of the marginal
Figure 16.5 3-D reconstruction of 64-slice CT angiogram illustrat-
arteries is based on their origin from the LCx; that is, first ing the LAD artery and an obtuse marginal branch in a left anterior
marginal, second marginal, and so forth. There is considerable oblique view with cranial angulation. Abbreviation: LAD, left anterior
anatomic variation in the number and size of the marginal descending. Source: Courtesy of Dr John Rumberger, Princeton
arteries, but there is at least one marginal branch present in Longevity Center.
most individuals (Fig. 16.5). When a single large marginal
branch is present, there is usually extensive secondary and
tertiary branching of the vessel. The most common pattern is to
have two or three marginal branches of similar size with less
extensive secondary branching. There is also a reciprocal rela-
tionship between the diagonal arteries and circumflex marginal
arteries. For example, when there are only a few relatively
small and short diagonals, there should be large marginal
branches of the LCx, which course quite anterior to supply a
portion of the anterolateral wall. Failure to see this pattern may
be a hint that there is a flush occlusion of a vessel and an area of
nonperfused myocardium. If a large ramus branch is present,
then the proximal diagonal branches are smaller and few in
number and supply an area close to the LAD. The ramus
supplies the next more lateral segment of the LV (LV wall
and then the LCx marginal branches pick up the supply as one
approaches the posterolateral portion of the left ventricular free
wall.
Right Coronary Artery

The RCA courses in the right atrioventricular groove and circum-
scribes the tricuspid valve annulus (Fig. 16.6). The first branch
arising from the RCA is frequently the conus or infundibular
branch, which courses anteriorly to supply the muscular right
ventricular outflow tract or infundibulum. Alternatively, in 30%
to 50% of angiograms the conus artery may arise from a separate
Figure 16.4 3-D reconstruction of 64-slice CT angiogram illustrat- ostium close to the RCA ostium or from a common aortic ostium
ing the coronary tree in a left posterior oblique projection. Note the shared with the RCA (19,28). Another important proximal
circumflex running posteriorly under the left atrial appendage to branch of the RCA is the sinus node artery. The reported
reach the left atrioventricular groove. Abbreviations: PA, pulmonary frequency of the sinus node artery arising from the proximal
artery; LAD, left anterior descending; OMB, obtuse marginal branch; RCA 50% to 70% with a dual blood supply from both the RCA
LA, left atrium; LCx, left circumflex. Source: Courtesy of Dr John and LCx in 3% (29,30). In its midportion, the RCA provides one
Rumberger, Princeton Longevity Center. or more branches that supply flow to the right ventricular free
wall (right ventricular marginal branches). The RCA also
[2/7/010/8:36:2] [173–193]
dilation. Therefore, the pressure drop along these conduit or

conductive arteries is negligible in the absence of a coronary
stenosis. The diameter of the epicardial coronary arteries, as
viewed by angiography, ranges from 1 to 5 mm, but arteries as
small as 500 mm can still be found on the epicardial surface.
The next component in series is the prearteriole, which
range in size from 100 to 500 mm. Because of their location and
wall thickness, these vessels are not directly affected by prod-
ucts of myocardial metabolism. Their role is to maintain pres-
sure at the origin of the arterioles within a narrow range in
response to changes in epicardial coronary perfusion pressure
and flow, a function that is often referred to as the autoregu-
lation of coronary blood flow.
The final component is the network of intramural arterio-
les, which have diameters less than 100 mm. The arteriole wall
consists of an endothelial layer facing the blood surrounded by
a layer of circumferentially oriented smooth muscle cells. These
are encased by connective tissue containing a rich plexus of
sympathetic and parasympathetic fibers. The smooth muscle
cells are easily able to constrict the lumen of an arteriole and
frequently do under physiologic and pathologic stimuli. Their
role is to regulate myocardial blood supply to match myocar-
dial oxygen consumption. This function is especially marked at
Figure 16.6 3-D reconstruction of 64-slice CT angiogram illustrat- their junction with the capillaries and blood passage into the
ing the right coronary artery in a right anterior oblique view. Abbre- capillaries is carefully controlled. The arterioles have a high
viations: LAD, left anterior descending artery; RCA, right coronary resting tone and dilate in response to metabolites released by
artery; RV, right ventricle; MPA, main pulmonary artery. Source: the myocardium as a result of an increase in oxygen consump-
Courtesy of Dr John Rumberger, Princeton Longevity Center. tion (33). Blood volume can be increased by 200% or more over
resting values in many capillary beds by the relaxation of the
arteriolar constrictors. Therefore, by regulating the resistances
in the prearterioles and arterioles, blood flow is matched with
supplies blood to the atria with a highly variable pattern of small oxygen requirements in the coronary circulation (34).
branches. The anatomy of the distal RCA is quite variable. In 50% The arterioles branch into numerous capillaries that lie
to 60% of individuals who have a right dominant pattern, the adjacent to the cardiac myocytes. The capillaries measure 10 to
RCA gives rise to the PDA at or near the crux. In the remaining 15 mm in diameter and are the site of metabolic exchange. A
individuals with right dominance, the PDA originates more high capillary to cardiomyocyte ratio and short diffusion dis-
proximally either at the acute margin (13%) or between the tances ensure adequate oxygen delivery to the myocytes and
acute margin and the crux (19%) (29). An anomalous origin of removal of metabolic waste products from the cells. Capillaries
the PDA from the first septal perforator without ischemic com- are composed of a single layer of endothelium with surround-
plications has also been reported (31). ing basement membrane and an incomplete layer of pericytes.
The anterolateral papillary muscle more frequently has a Pericytes, also known as Rouget cells or mural cells, are
dual blood supply from diagonal branches of the LAD and mesenchymal-like cells, associated with the walls of small
marginal branches of the LCx. Accordingly, it is more resistant blood vessels. As a relatively undifferentiated cell, they support
to developing ischemic dysfunction. In contrast, the poster- the small vessel, but can differentiate into a fibroblast, smooth
omedial papillary muscle is usually supplied only by the PDA muscle cell, or macrophage. They are important in angiogenesis
making it more vulnerable to ischemic dysfunction. The clinical and have been implicated in blood flow regulation at the
significance of this is that a myocardial infarction involving the capillary level. Capillaries are also surrounded by a loosely
PDA is more likely to cause mitral regurgitation (32). formed adventitia of collagen, elastic fibers and matrix.
Just as the epicardial coronary arteries are affected by
The Coronary Microcirculation atherosclerosis, the coronary microcirculation is affected in a
The left and right coronary arteries and their major branches lie variety of systemic and cardiac disorders. These consist of
on the surface of the heart, and thus are referred to as the functional alterations, involving changes in the responsiveness
epicardial coronary vessels. However, the entire coronary arte- of the coronary microvasculature, and structural changes
rial system is composed of three distinct components, which involving alterations in the number and diameter of the coro-
have different physiologic functions. A detailed description of nary microvessels. A proposed a classification of the patho-
coronary microvascular anatomy and function is beyond the genic mechanisms of coronary microvascular dysfunction is
scope of this chapter, but an overview is important to the overall shown in Table 16.1 (35).
understanding of the coronary circulation. Although there are
three distinct components of the coronary circulation, these are
not demarcated by distinct anatomic borders. The most proxi- Coronary Venous Anatomy
mal component comprises the large epicardial coronary arteries. The capillaries terminate in small venules, vessels with a diam-
These function primarily as conduits offering little resistance to eter of 15 mm or more. Other than size, the morphologic change
coronary blood flow, but are subject to some flow-mediated from capillary to venule is not very distinctive. Small venules
[2/7/010/8:36:2] [173–193]
Table 16.1 Pathogenic Mechanisms of Coronary Microvascular Dysfunction

Type Alteration Cause
Structural Luminal obstruction Microembolization in acute coronary syndromes or after recanalization
Vascular wall infiltration Infiltrative heart disease (e.g., Anderson-Fabry cardiomyopathy)
Vascular remodeling Hypertrophic cardiomyopathy, arterial hypertension
Vascular rarefaction Aortic stenosis, arterial hypertension
Perivascular fibrosis Aortic stenosis, arterial hypertension
Functional Endothelial dysfunction Smoking, hyperlipidemia, diabetes
Dysfunction of smooth muscle cell Hypertrophic cardiomyopathy, arterial hypertension
Autonomic dysfunction Coronary recanalization
Extravascular Extramural compression Aortic stenosis, hypertrophic cardiomyopathy, arterial hypertension
Reduction in diastolic perfusion time Aortic stenosis
have an endothelial lining, a surrounding basement lining, and cardiac vein terminates in the coronary sinus, a junction defined
collagen connections from the basement membrane to the by the presence of the left atrial oblique vein. This transition
surrounding matrix. However, the venules are unlike capilla- point is usually marked by the presence of intravenous valves,
ries in that vasoactive compounds like histamine, various which can obstruct catheter and pacemaker lead placement.
kinins and serotonin can affect the separation of the tight Another fairly consistent branch is the middle cardiac vein,
junctions of the endothelial cells resulting in the leakage of which runs in the posterior interventricular groove, parallel to
large molecular weight substances. Veins of about 0.5 mm in the posterior descending coronary artery. Of all of the branches
diameter begin to acquire a muscular coat and eventually form of the coronary venous system, the great cardiac and middle
small epicardial veins that run in parallel with the visible cardiac veins are the two most consistently present branches
epicardial arterial branches. seen in more than 90% of individuals (39,40). However, unlike
There is considerable variability in the cardiac venous the middle cardiac vein, the great cardiac vein varies consider-
anatomy, but there are some consistent venous structures ably in its course (41). Lateral and posterior venous branches
(Fig. 16.7) (36). The most notable of these is the great (or together are seen in <50% of human hearts.
anterior) cardiac vein, which begins at the apex of the heart The coronary sinus is the most constant feature of the
and ascends along the anterior interventricular groove parallel cardiac venous system, although several congenital anomalies
to the LAD to the base of the ventricles. It connects with have been described (42–44). The coronary sinus lies in the
diagonal veins draining the lateral and anterolateral portion of atrioventricular groove on the posterior surface of the heart and
the LV and turns posterior at the left atrioventricular groove receives veins from the lateral wall, which are referred to as
wrapping around the left side of the heart parallel to the LCx marginal veins. Although the coronary sinus invariably lies in
coronary artery (38). In addition to several smaller tributaries the atrioventricular groove, its branches and their locations are
from the left atrium and ventricles, the great cardiac vein far more variable than those of the coronary arterial system
receives two main branches, the large left marginal vein along (43). The coronary sinus opens into the right atrium, poster-
the lateral border of the heart and the posterior left ventricular omedially, just superior to the septal leaflet of the tricuspid
branch (also known as the posterolateral branch). The great valve. At the ostium of the coronary sinus is the thebesian
Figure 16.7 Cardiac venous anatomy shown in an anterior (left) and posterior (right) views. Abbreviations: LA, left atrium; RA, right atrium;
RV, right ventricle; LV, left ventricle; SVC, superior vena cava; IVC, inferior vena cava. Source: From Ref. 37, Mayo Foundation for Medical
Education and Research.
[2/7/010/8:36:2] [173–193]
valve, a semicircular fold of the lining membrane of the atrium. the ventricles, myogenic cells proliferate and form extensive
The valve may vary in size, or be completely absent and may trabeculae. In the posterior regions, however, proliferating
prevent regurgitation of blood into the coronary sinus during myocytes do not form extensive trabeculations and this region
contraction of the atrium. The valve can also hinder cannula- eventually becomes the atria (48). By the end of the first 24 days
tion of the coronary sinus os (45). The coronary sinus and its in humans, the heart has an endocardium and a myocardium,
tributaries drain approximately two-thirds of the left ventricu- but lacks an epicardium and there are no rudimentary blood
lar myocardium, but they do not drain the superior part of the vessels (49,50).
interventricular septum, the right atrium and ventricle or the None of the cells that form the coronary system actually
myocardium of the roof of the left atrium. These are drained by arises from within the heart. All the cells that form the coronary
the right cardiac (or anterior cardiac) venous system. These system come from outside the heart and then differentiate into
veins originate on the anterolateral surface of the right ventricle blood vessels only when they are within the heart. All of this
and drain directly into the right atrium. In addition, there are occurs without interaction with the blood coursing through the
also thebesian veins (venae cordis minimae), which drain primitive heart lumen (50). There are at least three major
directly into the cardiac chambers. They are more common anatomic components that are important in the development
on the interventricular and interauricular septum, particularly of the coronary arteries (19). First, the myocardial sinusoids are
on the right side of the heart, but prominent thebesian veins are an elongation of the trabeculae into the primitive loosely
occasionally seen entering the LV. Other variable features of the packed myocardium and thus communicate with the heart
coronary venous anatomy include the presence of ostial valves cavities. These sinusoids are the earliest sites of metabolic
of the cardiac veins (Vieussens valves), interbranch collateral- exchange whereby the developing myocardium is nourished.
ization, and intramural versus epicardial course. As the myocardium becomes more compact, the sinusoids
The knowledge of anatomical positioning of the cardiac disappear, but persistence of these sinusoids may lead to cor-
venous structures is becoming increasingly important in the onary artery-cameral fistulaes. Second, by about 32 days after
field of cardiac electrophysiology for positioning of mapping ovulation a separate in situ vascular network begins to develop.
catheters and left ventricular pacing leads. Implantation of the This primitive network of arteries, veins, and capillaries may
coronary sinus lead usually involves the lateral and posterior have connections with other mediastinal vessels, which can be
branches, which are quite variable in their number, tortuosity, a source for coronary artery fistulae. Third, as the coronary
dimensions, and angulation with respect to the main trunk of artery network evolves, endothelial buds arise from the base of
the atrioventricular venous ring (46). the truncus arteriosus as septation is occurring. It is still unclear
if there are initially only two buds or six buds, one from each
potential cusp of the aortic and pulmonary sinuses with later
VARIATIONS IN CORONARY ANATOMY involution of all but two buds. These buds grow and, after
Although there is considerable heterogeneity in coronary anat-
septation is complete, fuse with the developing in situ coronary
omy among individuals, there is greater consistency in the
artery network to form the coronary artery system (Fig. 16.8).
regions of the heart generally supplied by the different coro-
nary arteries (Table 16.2). This anatomic distribution is impor-
tant because these cardiac regions are assessed by 12-lead
electrocardiograms to help localize ischemic or infarcted seg-
ments of the myocardium. This in turn, can be loosely associ-
ated with specific coronary vessels, but because of vessel
heterogeneity, the specific vessel involvement requires verifi-
cation by coronary angiography or other imaging techniques.
Embryology of the Coronary Arteries

Although a detailed description of the formation of the coro-
nary vasculature is beyond the scope of this chapter, some
fundamental principles are important for understanding some
variations in coronary anatomy and coronary anomalies. Dur-
ing the earliest stages of cardiogenesis the heart is formed as an
endothelial tube within a muscular tube (47). The lateral plate
mesoderm is the source of both the endothelium and the
muscle layers and these are the only two cardiac cell types
generated. Anteriorly, in the region that will eventually become
Table 16.2 Myocardial Regions Associated with Epicardial Coronary

Arteries Figure 16.8 Schematic representation of the basic components
Region Coronary artery most likely associated involved in the embryogenesis of coronary arteries. Aorta (AO) and
pulmonary (PA) trunks are shown at completion of septation; coro-
Inferior Right coronary artery or distal circumflex nary buds (3a, 3b, 3c, 3d) emerge from semilunar sinuses. Rudi-
Anteroseptal LAD ments of right (Ca), circumflex (Cb), and left anterior descending
Anteroapical Distal LAD (Cc), coronary arteries are shown as isolated in situ vascular net-
Anterolateral Diagonal branches of LAD or obtuse marginal works. At this stage sinusoids (Sn) are site of metabolic exchanges
arteries between intracavitary blood and cardiac jelly. Source: From Ref. 19.
Abbreviation: LAD, left anterior descending.
[2/7/010/8:36:2] [173–193]
Although the coronary ostia are formed early, the distal coro- when standard angiographic catheters or manipulations fail to
nary branching pattern remains variable until the cardiac cannulate the artery. Similar to a high origin of the RCA, a high
chambers are more fully developed. Abnormal involution (in takeoff of the left main is not generally felt to have any clinical
the case of six initial buds), bud position, or septation of the significance, but there is one older report of showing morpho-
truncus arteriosus are all factors that may contribute to the logic evidence of chronic ischemia and left ventricular scarring
development of coronary artery anomalies. in a patient with a high takeoff of the left main (55). A very high
origin of the left main should be noted if cardiac surgery is
planned so as to avoid accidentally crossclamping or trans-
Variations in Coronary Anatomy secting the vessel during surgery (56). Similar to the RCA, the
of Minimal Clinical Significance origin of the left coronary has been rarely observed from the
Right Coronary Artery noncoronary cusp, but has never been associated with symp-
Normally, there are two main coronary ostia but several com- toms or complications in this location (53).
mon variations may occur. As noted earlier, in 30% to 50% of The left coronary ostium is usually single and is the
angiograms the conus branch of the RCA may arise separately origin of the left main coronary artery. The most frequent
from the right sinus rather than its usual position as a branch of minor variation observed is absence of the left coronary,
the proximal RCA (19,28). Two right conal branches have also which then results in the LAD and LCx arising directly from
been reported. A separate conus branch usually has a diameter the aortic root as separate vessels. The vessels are otherwise
at the ostium varying from 0.5 to 1.9 mm and may give rise to normal in their respective distributions. The incidence of truly
preventricular and ventricular branches, which nourish more separate ostia varied from 0.5% to 8% in otherwise normal
than just the infundibular myocardium of the outflow tract. In hearts (57). The wide variation in the occurrence of this finding
patients with occlusion of the LAD or RCA, the conus artery in the literature is, in part, related to the absence of a consistent
often serves as a principal source of collateral circulation. In definition for this entity. If defined as two separate ostia from
such patients it is important to visualize the conus artery well the aorta, the incidence is probably less than 1%. However, if it
to adequately visualize the collateralized vessels (51). is defined simply as the absence of a proper left main trunk, the
The location of the RCA ostium is more variable than the incidence is more than 1%. Rather than two distinctly separate
left ostium. This variability makes it difficult to define what vessels, the more common observation is a single ostium, which
exactly a normal location is, but an origin below the aortic ring is shared by the origins of the LAD and LCx.
is definitely “low” and more than 1 cm above the sinotubular
ridge is definitely a “high” takeoff. A high origin of the RCA is
generally thought to be of no clinical consequence, but sudden Variation in Coronary Size
death was reported in an amateur athlete with a high origin of Various techniques have been used to determine the size of
the RCA plus small, hypoplastic right and LCx arteries (52). normal coronary arteries and it is now generally appreciated
Furthermore, instead of the usual location in the middle of the that size measurements from autopsy specimens do not corre-
sinus the RCA ostium can be located closer to the aortic valve late well with measurements made in vivo (58,59). Some of this
commissures. This causes a slight alteration in the usual course discrepancy is due to the techniques used for fixation of the
of the vessel, but has no other clinical significance. The RCA autopsy specimens (60). In practical terms, since atherosclerosis
may arise from the posterior (noncoronary) sinus but this is a is a diffuse process it can sometimes be difficult to clinically
very rare anomaly, which has never been associated with distinguish a diffusely diseased segment from one that is just
symptoms or complications (53). Finally, a RCA arising from normally small in diameter. Despite a need for these data,
the midsegment of the LAD has been described and presumed relatively little information exists on lumen size variations in
not to cause ischemic complications (54). normal coronary arteries. Coronary artery diameters at multi-
ple locations were determined in carefully selected normal
Left Coronary Artery coronary arteries using computer-based quantitative measure-
The left coronary orifice usually arises from the center of the ments by Dodge and colleagues (Table 16.3) (61). Using these
left sinus of Valsalva just above the free margin of the aortic and other measurements (62) it is possible to group the size of
valve leaflet. Malposition of the left coronary ostium either high certain segments of normal coronaries, but the wide standard
or low in the sinus or near the aortic valve commissures occurs deviation makes the application of these to an individual
less often compared with the RCA, but should be suspected patient of less value. Nevertheless, some associations are
Table 16.3 Diameter Measurements of the Main Coronary Arteries in Normal Men
Location Right dominant (mm) Left dominant (mm)
Left main: midway between ostium and bifurcation 4.5 0.5 4.6 0.4
LAD (1st segment): midway between its origin and 1st septal 3.6 0.5 3.7 0.2
LAD (3rd segment): midway between the 3rd septal and apex 1.7 0.5 2.0 0.3
LCx (1st segment): midway between its origin and the 1st marginal 3.4 0.5 4.2 0.6
LCx (3rd segment): midway between the 1st marginal and most distal marginal 1.6 0.6 3.2 0.5
RCA (1st segment): midway between its origin and the 1st acute marginal 3.9 0.6 2.8 0.5
RCA (3rd segment): midway between the 3rd acute marginal (if present) and 3.1 0.5 1.1 0.4
posterior descending
Abbreviations: LAD, left anterior descending; LCx, left circumflex artery; RCA, right coronary artery.
[2/7/010/8:36:2] [173–193]
noted. For example, the RCA or LCx is significantly larger may have systolic compression, but these are referred to as
when it is the dominant vessel, but the PDA is similar in size myocardial loops rather than classic myocardial bridges. As
whether it arises from the RCA or LCx. One thing their data did another variant, arteries such as an obtuse marginal branch or
show was that coronary arterial diameter in women was about ramus located over the free wall of the LV may dive into the
9 8% smaller than in men, even after normalization for body myocardium and not resurface.
surface area. The smaller arterial size of the left main and Myocardial bridges and tunneled arteries have long been
proximal LAD in women was confirmed in a more recent recognized clinically and are generally felt to be a benign inci-
study using intravascular ultrasound (63). Other studies have dental finding (66). Indeed, in most patients myocardial bridges
shown a good correlation between the lumen area and of a have little clinical significance as the majority of coronary flow
coronary artery along its length and the corresponding in the left coronary occurs during diastole when there is no
summed distal branch lengths and regional myocardial mass compression (70). However, there are scattered reports in which
in patients with and without coronary disease, thus the greater typical angina with anterior wall ischemia during tachycardia,
amount of distal territory supplied the larger the caliber of the myocardial perfusion defects, acute myocardial infarction
artery lumen (64). abnormal ventricular repolarization and sudden cardiac death
have all been attributed to myocardial bridges (71–75). More
Absent Circumflex or RCA elegant studies using intravascular ultrasound and Doppler
At the extreme of variations in coronary size are the rare flow velocity have documented disturbed intracoronary hemo-
anomalies of an absent circumflex artery or RCA. In this case dynamics in both symptomatic and asymptomatic patients with
of an absent circumflex, the RCA is a very large vessel (super- myocardial bridging in the midportion of the LAD (76,77). In
dominant), which crosses the crux and ascends in the atrioven- symptomatic patients, treatment by either stent placement or
tricular groove on the left to supply the posterior and lateral surgical transsection of the muscle bridge has relieved both
myocardium. In the case of an absent RCA, the circumflex symptoms and objective findings of myocardial ischemia
continues in the atrioventricular groove in the course typical of (77–79). Several anatomic studies have reported a “protective”
the normal RCA (65). The LAD is usually normal in its size and effect of myocardial bridging on the development of atheroscle-
distribution. Neither of these anomalies appears to be of clinical rosis. The mechanism of this effect is unknown, but has been
significance in the absence of coronary artery disease. postulated to be from a reduction in systolic wall stress in the
tunneled segment. In humans, myocardial bridges may slightly
increase the occurrence of proximal atherosclerosis while pro-
Variations in the Course of Coronary Arteries tecting the bridged segment and the distal artery. Careful
Myocardial Bridging autopsy studies have shown that when myocardial bridging is
As noted previously, minor variations in the course of a coro- present, intimal thickening and macroscopic raised atheroscler-
nary artery may occur if the ostium is malpositioned within the otic plaques are increased just proximal to the bridge (80).
aortic cusp, but these are usually of no clinical importance. One
of the most common variations in the course of a coronary Crossing Epicardial Coronary Arteries
artery occurs when a segment of the conduit artery dips into As the name implies, two coronaries artery branches may rarely
the myocardium resulting in the overlying myocardium com- cross over each other on the epicardial surface. The true inci-
pressing the artery during systole (66). The muscle overlying dence of this minor variation in coronary anatomy is unknown
the intramyocardial segment of the coronary artery is termed a and it is not believed to have any functional significance (81,82).
myocardial bridge and the artery coursing through the myo- Crossing of two right ventricular branches, an acute marginal
cardium is called a tunneled artery. The most frequent site of branch of the RCA with the RCA in the atrioventricular groove,
bridging in man is the midsegment of the LAD. A typical a diagonal and circumflex marginal branch, two obtuse mar-
muscular bridge in this segment is 10 to 20 mm long and 2 to ginal branches of the circumflex and the LAD and a diagonal
4 mm thick, but segments up to 50 mm in length have been have all been reported.
observed (67). Muscular bridges have been identified over
diagonal arteries, the left main, LCx, marginal branches and Intercoronary Continuity
the RCA. Autopsy studies show a higher incidence of myocar- Intercoronary artery continuity or “coronary arcade” is a rare
dial bridges than demonstrated by angiography. For example, variant of the coronary circulation. The true incidence is
angiographic studies show a prevalence of bridging varying unknown, but in one report it was seen in 0.02% of nearly
from 0.5% to 7.5% of studies whereas anatomic studies show a 10,000 coronary angiograms (83). Arterial continuities exist in
prevalence as high as 60% in the LAD and 6% to 50% in other other areas such as the superficial volar arch in the hand,
vessels (67–69). Factors such as the length of the tunneled intestinal branches of the superior mesenteric artery and the
segment, the degree of systolic compression and the heart circle of Willis. Communications between the distal LCx and
rate have all been postulated to explain the difference in prev- the distal RCA in the posterior atrioventricular goove and
alence between angiographic and autopsy studies. In addition, between the LAD and PDA in the distal interventricular groove
for systolic narrowing to occur the external muscular compres- have both been observed (Fig. 16.9). These connections occur in
sive force must exceed the arterial pressure and the intrinsic the absence of obstructive coronary disease and pathologically
arterial wall stiffness. During angiography, the increased intra- distinct from coronary collaterals (84). Compared with collat-
luminal pressure related to the contrast injection may act to erals, these connections have a well-defined muscular layer, are
diminish the appearance of systolic compression and thus the larger in diameter (>1 mm) and extramural.
appreciation of a myocardial bridge. Some anatomic variation
also exists. Arteries located in the atrioventricular groove Small Coronary Artery Fistulas
(proximal RCA and LCx) may be surrounded by scattered A coronary artery fistula is an abnormal communication
muscular fibers continuous with the atrial myocardium and between an epicardial coronary artery and a cardiac chamber,
[2/7/010/8:36:2] [173–193]
Table 16.4 Incidence of Coronary Anomalies Detected by

Angiography
Author (reference number) Patients (n) Incidence (%)
Yamanaka O (94) 126,595 1.30
Wilkins CE et al. (93) 10,661 0.78
Kimbiris D et al. (91) 7,000 0.64
Baltaxe HA (87) 1,000 0.90
Chaitman BR et al. (88) 3,750 0.83
Engel HJ (90) 4,250 1.20
Cieslinski G (89) 4,016 0.97
consider these to be variants of normal. In general, congenital

coronary artery anomalies in the absence of other cardiac
congenital anomalies have been described in approximately
1% of patients who undergo coronary angiography and
approximately 0.3% of patients at autopsy (Table 16.4) (87–94).
There are several congenital syndromes that are associ-
Figure 16.9 Coronary angiogram of the right coronary artery in a ated with coronary anomalies. Patients with Williams syn-
left anterior oblique view with cranial angulation. The catheter is in drome (elfin facies, infantile hypercalcemia, hypoplastic teeth)
the right coronary artery and a forceful injection used to visualize the may have coronary ostial narrowing as a component of supra-
entire coronary circulation with reflux of contrast into the aorta. valvar aortic stenosis, which is characteristic of this disease.
Abbreviations: RCA, right coronary artery; LAD, left anterior Patients with congenital aortic valve disease commonly have
descending; PDA, posterior descending artery; LCx, left circumflex;
variants in coronary ostial origin. Coronary anomalies may be
c, communicating artery.
commonly associated with other congenital cardiac malforma-
tions, most notably, transposition of the great arteries, tetralogy
of Fallot and different forms of pulmonary atresia (95,96).
major vessel or other vascular structure. Small coronary artery
fistulas occur in 0.1% to 0.2% of patients undergoing coronary
angiography and usually drain into a single cardiac structure Coronary Anomalies Not Associated
(85). One of the most commonly observed small fistulas in with Ischemic Complications
adults is a communication between the LAD and the common Origin of the LCx from the Right Coronary Sinus
pulmonary artery. Small fistulas are usually not associated with This anomaly is said to be the most common anomaly of
continuous murmurs or detectable intracardiac shunts and have coronary arterial origin in adults with an incidence of about
a benign course. Small numbers of patients have undergone 0.3% when assessed by coronary angiography and autopsy
serial angiographic studies demonstrating no increase in fistula (89,93,97). The aberrant LCx can arise from either the proximal
size over time and patients have been followed for up to 11 years segment of the RCA or from a separate ostium near the origin
without complications attributable to their fistula (86). The of the RCA (Fig. 16.10). The LAD and RCA are normal in their
detection of small fistulas in the elderly confirms their benign distribution and the anomalous LCx invariably courses poste-
nature. The majority of coronary fistulas are congenital and arise rior to the aorta and then to its normal distribution over the
from two defects in coronary artery embryogenesis. Fistulas can lateral wall of the heart (Fig. 16.11). Most patients have no other
arise from failure of the embryonal intramyocardial sinusoids to associated anomalies. This anomaly should be suspected when
obliterate resulting in a gradual enlargement of one of these angiography of the left coronary shows an unusually long,
channels or from enlargement of a thebesian veins (85). Alter- nonbranching proximal segment and no obvious LCx vessel
natively as in the case of fistulas involving the pulmonary perfusing the lateral wall (98). This anomaly can be missed
artery, it may evolve from persistence of a vascular bud that during angiography of the RCA if the tip of the catheter is
remains attached to the pulmonary artery as it separates from beyond the origin of the anomalous vessel and the force of the
the aorta early in embryogenesis. Patients with small asympto- injection is inadequate to fill the aberrant LCx with the reflux of
matic fistulas require no special treatment and can simply be contrast. It can also be suspected if a “dot sign” is seen in a right
followed clinically. Small fistulas can also be acquired and have anterior oblique left ventriculogram or supravalvular aorta-
been reported secondary to deceleration injuries, coronary gram (Fig. 16.12). This anomaly has traditionally been consid-
angioplasty, repeated myocardial biopsies in heart transplant ered of little clinical significance unless the operator incorrectly
patients, pacemaker leads and after cardiac surgery. assumes the LCx is occluded or an important stenosis in the
aberrant artery is not visualized. This is true in the majority of
cases, but there are a few case reports where ischemic compli-
CORONARY ARTERY ANOMALIES cations have been associated with this anomaly (99,100). Fur-
The exact incidence of coronary anomalies is unknown and thermore, it has been reported that the stenosis severity in this
the reported incidence varies depending on the methods used anomaly is greater than in control subjects matched for age,
to detect the anomalies, the population assessed and what is gender, symptoms and degree of atherosclerosis in nonanom-
included as a coronary anomaly as opposed to a variant of alous coronary arteries (101). Compression of the anomalous
normal. Some sources classify several of the anatomic varia- LCx has been reported from the fixation rings of prosthetic
tions described above as anomalies while other sources simply valve (102).
[2/7/010/8:36:2] [173–193]
Figure 16.12 The “dot” sign is seen on a right anterior oblique left
ventriculogram or supravalvular aortagram and represents the
anomalous circumflex in its course posterior to the aorta (Ao).
Figure 16.10 64-slice CT coronary angiogram of LCx arising from
the RCA and traversing posterior to aorta. Abbreviations: LCx, left
circumflex; RCA, right coronary artery. Source: Courtesy of Dr John
Rumberger, Princeton Longevity Center.
Origin of a Coronary from the Posterior Sinus of Valsalva
Either the RCA or the left main coronary artery can arise from
the posterior sinus of Valsalva (the noncoronary sinus). Both of
these variants are considered extremely rare, although the RCA
from this location is said to be more frequent (53). Origin of the
left main from the posterior cusp has not been reported in the
absence of other anomalies of the heart and great vessels and
the simultaneous origin of both arteries from this location has
never been reported (53).
Certain Variations of a Single Coronary Artery

There are many variations of a single coronary artery and
several anatomic classifications have been proposed (103,104).
Lipton’s classification now seems to be the most frequently
used (103). In this classification, the letters “R” and “L” indicate
whether the single artery arises from the right or left aortic
sinus and the letters “A,” “B,” and “P” signify whether the
“transverse” branch courses anterior, between or posterior to
the great vessels, respectively (Fig. 16.13). The associated num-
ber in this classification indicates the number of major branches
somewhat mimicking the RCA, LCx, and LAD. When found in
a younger age group (<20 years) a single coronary artery is
frequently associated with other anomalies of the heart and
great vessels, but when identified later in life associated
anomalies are infrequent (53). From the cases reported, there
is a slight male predominance (male-to-female ratio 1.4:1) with
Figure 16.11 Anomalous LCx coronary artery. Injection of the a similar frequency of arteries arising from the right or left
RCA in the left anterior oblique projection slight cranial angulation sinuses of Valsalva. Certain types of single coronary artery
shows the anomalous LCx with a typical course posterior to the have not been associated with clinical complications. In gen-
aorta. Abbreviations: LCx, left circumflex; RCA, right coronary eral, these are the varieties where no vessel courses between the
artery. Source: Courtesy of Cardiac Catheterization Laboratory, great vessels (Fig. 16.13, top). However, should coronary athe-
Scott & White Healthcare. rosclerosis develop in the main trunk of a single artery, the
complications of atherosclerosis can be devastating.
[2/7/010/8:36:2] [173–193]
Figure 16.13 Classification of single coronary artery anomalies. In this classification, the letters “R” and “L” indicate whether the single artery
arises from the right or left aortic sinus and the letters “A,” “B,” and “P” signify whether the “transverse” branch courses anterior, between or
posterior to the great vessels, respectively. The associated number in this classification indicates the number of major branches somewhat
mimicking the right, left circumflex and left anterior descending arteries. In the diagram, the great vessels from front to back are the pulmonary
artery, aorta, and superior vena cava. Those anomalies not typically associated with complications (benign) are shown on the top and those
potentially associated with ischemia are on the bottom. Source: Adapted from Ref. 103.
Coronary Anomalies Associated monary artery pressure are possible explanations. If a substan-
with Clinical Complications tial collateral circulation between the RCA and the anomalous
Origin of One or More Arteries from the Pulmonary Trunk left coronary artery persists survival into adulthood is possible
Coronary arteries arising from the pulmonary artery are one of with a left-to-right shunt from the RCA to the pulmonary
the most serious congenital coronary artery anomalies. It fre- artery.
quently results in death during infancy and only a few indi- In the most common infantile presentation of this anom-
viduals survive to adulthood (105). Either the left or right aly, collaterals are inadequate to support the LV and the child
coronary artery, both major coronary arteries, or rarely, the presents with tachypnea, cough, wheezing, pallor and cyanosis.
LAD or LCx may originate from the pulmonary artery. Occa- Signs of myocardial ischemia and infarction with the associated
sionally, origin of the conus artery from the pulmonary artery complications of aneurysm formation, mitral regurgitation and
has been noted. The most frequent variety of this anomaly is congestive heart failure are present (105). Diagnostic testing
origin of the left coronary from the pulmonary artery and a confirms a pattern of anterior infarction on the electrocardio-
RCA arising from the usual position in the aorta with a normal gram in many with important mitral regurgitation demon-
course (Fig. 16.14). In infants this anomaly has been referred to strated by echocardiography and Doppler. Angiography is
as the Bland-White-Garland syndrome (107). Children with this frequently necessary to confirm the diagnosis. Prompt surgical
anomaly usually present between 8 and 16 weeks after birth, therapy is indicated as survival without surgery is <20% in
but adult presentation is occasionally seen (108). During fetal symptomatic infants (109). Some reports suggest that up to 20%
life, the systemic pressure and oxygenization in the pulmonary of patients with this anomaly may survive into adulthood and
artery provide adequate perfusion to the LV. However, within remain asymptomatic or have a late presentation when they
days after birth the pressure and oxygen content of the pulmo- develop mitral regurgitation, angina, or CHF (110). Sudden
nary artery decline so that the left coronary is underperfused. It death is a complication in both infants and adults. The clinical
is not clear why the clinical presentation is delayed for several course of the patient tends to be more favorable if extensive
weeks but persistence of collaterals and a delayed fall in pul- collateral circulation exists.
[2/7/010/8:36:2] [173–193]
Figure 16.14 Two-dimensional echocardiographic image in a short

axis view of an infant who has an anomalous origin of the LCA from
the PA. The anomalous LCA can be seen arising from the PA,
whereas the mildly dilated RCA arises from its normal position off the Figure 16.15 64-slice CT coronary angiogram of the RCA arising
aorta (AO). Abbreviations: LCA, left coronary artery; PA, pulmonary from the left coronary cusp and coursing posterior to the PA.
artery; RCA, right coronary artery. Source: From Ref. 106. Abbreviations: PA, pulmonary artery; RCA, right coronary artery;
LA, left atrium; LM, left main coronary artery. Source: Courtesy of
Dr John Rumberger, Princeton Longevity Center.
Anomalous origin to the RCA from the pulmonary artery

is much less common and has fewer clinical consequences and
the same is true for a conus branch arising from the pulmonary
trunk. If the LAD alone arises from the pulmonary artery the
presentation in both children and adults is often similar to that
of the left main in this position. Depending on the extent of
collaterals, symptoms and signs of myocardial ischemia may
occur in infancy or be delayed until adulthood and surgical
repair is often necessary (53). In some cases of this anomaly,
only ligation of the anomalous LAD is performed while in other
cases, ligation and some form of bypass is used. Anomalous
origin of the circumflex alone has been reported in children
with other cardiac abnormalities, but no adults have been
reported with this anomaly (53).
Origin of Both Right and Left Coronary Arteries

from the Same Sinus of Valsalva
When either the RCA arises from the left sinus of Valsalva
(Fig. 16.15) or the left coronary arises from the right sinus of
Valsalva (Fig. 16.16), the anomalous vessel traverses the base of
the heart in one of four possible paths. The anomalous vessel
can pass Anterior to the pulmonary trunk (type A) (Fig. 16.16), Figure 16.16 3-D 64-slice CT coronary angiogram illustrating the LM
Between the aorta and pulmonary trunk (type B), through the arising from the RCA and traversing anterior to the aorta (Ao) and
Crista supraventricularis (within the ventricular septum posterior to the pulmonary artery. In this reconstructed image the
beneath the right ventricular infundibulum) (type C), or pos- pulmonary artery has been cut away and is not shown to allow
terior or Dorsal to the aorta (type D) (Fig. 16.17) (53,103). The visualization of the LM. Abbreviations: RCA, right coronary artery;
angiographic appearance of these various courses has been well LM, left main coronary artery. Source: Courtesy of Dr John Rumberger,
described (Fig. 16.18) (111). There is now agreement that the Princeton Longevity Center.
course of an anomalous coronary artery, rather than the actual
location of the coronary ostium, is a major discriminating factor
for the anomaly being benign or associated with clinical passes between the aorta and the pulmonary artery, or less
complications such as angina, ventricular arrhythmias, syncope commonly via a septal pathway. The mechanism of ischemia,
or sudden death (53,94). Adverse outcomes occur more fre- infarction or sudden death in this situation appears related to
quently when the anomalous coronary artery has a course that the shape of the coronary ostium of the anomalous vessel rather
[2/7/010/8:36:2] [173–193]
varieties have been recognized. The anomalous LAD can take a

path anterior to the pulmonary artery and this course is
frequently seen in association with tetralogy of Fallot. It is
important to identify this course in children undergoing repair
of the tetralogy of Fallot so as to avoid damage to this branch
during surgery. Although this course is usually not associated
with clinical complications in adults, isolated case reports of
ischemic complications exist (120–122). Ischemic complications
have also been reported to occur when the LAD courses
between the aorta and pulmonary artery (123,124) or has a
course within the interventricular septum (125). Certain varia-
tions of a single coronary artery are quite similar to a wrong-
sided coronary artery and are also associated with ischemic
complications (Fig. 16.13) (126).
Unfortunately, many patients with these anomalies pres-
ent with sudden cardiac death, usually occurring during or
immediately after intense athletic activity. In one large series of
cases presenting with sudden cardiac death and proven at
autopsy to have either the left or right coronary artery arising
from the wrong side and passing between the aorta and pul-
monary artery, only 45% had premonitory symptoms of chest
Figure 16.17 Anomalous left main coronary artery from the right pain or syncope (127). In those who had some type of cardio-
sinus of Valsalva. The anomalous vessel traverses the base of the vascular test before death, all tests were within normal limits,
heart in one of four possible paths: Anterior to the pulmonary trunk including 12-lead electrocardiograms, stress ECG with maximal
(type A), Between the aorta and pulmonary trunk (type B), through exercise and left ventricular wall motion and cardiac dimen-
the Crista supraventricularis (within the ventricular septum beneath sions by two-dimensional echocardiography. If there is a sus-
the right ventricular infundibulum) (type C), and posterior or Dorsal picion of this anomaly, visualization of the coronary arteries by
to the aorta (type D). Abbreviation: RCA, right coronary artery. computed tomographic angiography, magnetic resonance
Source: Adapted from Ref. 103. imaging or conventional angiography is usually necessary
(128,129).
than compression of the anomalous vessel by the aorta and

pulmonary artery during systole. Normally, the coronary ostia Coronary Atresia
are round to oval in shape, but in these anomalies the tranverse Atresia or hypoplasia of the left main or RCA is a very rare
course of the vessel to the opposite side of the heart results in cause of ischemia and myocardial infarction in infancy and
an acute-angle takeoff that converts the ostium into a slit-like early childhood (130). Survival to adulthood depends on the
shape (Fig. 16.19) (112,113). Clinical events, in particular sud- development of adequate collateral flow from the other main
den death, are usually seen during exertion in young individ- coronary artery. Atresia of a coronary artery can occur as an
uals in the absence of coronary atherosclerosis or other cardiac isolated lesion or in association with other congenital diseases
abnormalities. The increased cardiac output during exercise such as supravalvular aortic stenosis, homocystinuria, Frie-
dilates and stretches the aortic wall resulting in a further dreich’s ataxia, Hurler’s syndrome, progeria, and pulmonary
kinking and additional compromise of the slit-like opening atresia (131). Successful surgical repair is possible (132).
thereby causing a transient limitation of coronary blood flow.
A left main artery arising from the right sinus of Valsalva
and coursing between the aorta and pulmonary artery is the Large Coronary Artery Fistulas
most threatening variety in this group of anomalies (53,114). As described earlier, a coronary artery fistula is an abnormal
Complications have not been associated with a course of the left micro- or macrovascular connection between a coronary artery
coronary artery that is either behind the aorta or anterior to the and a cardiac chamber, vein, or another artery. Fistulas
pulmonary artery, but there are reports of ischemic complica- between a coronary artery and a cardiac chamber are also
tions when the course is through the upper septum called coronary-cameral fistulas and those between a coronary
(94,115,116). Although there are other anatomic possibilities, a artery and vein are termed coronary arteriovenous fistulas
right coronary artery arising from the left sinus of Valsalva or (Fig. 16.20). Small fistulas rarely are associated with clinical
the proximal left main almost invariably (>99%) will pass findings or complications and are usually detected as an inci-
between the aorta and pulmonary artery (94). The clinical dental finding on angiography (85). About half of the patients
course in these patients is variable with some having no with a coronary artery fistula are asymptomatic, but the clinical
symptoms or evidence of cardiac dysfunction and others hav- presentation in those with large fistulas depends on the type of
ing complications including angina, myocardial infarction, syn- fistula, shunt volume, site of the shunt and presence of other
cope and sudden death (117,118). There is some suggestion that cardiac conditions. Detection of a continuous murmur, dyspnea
a familial clustering of anomalous origin of a coronary artery with exertion, fatigue, congestive heart failure, arrhythmias,
from the wrong aortic sinus with an intra-arterial course may pulmonary hypertension, infective endocarditis, or myocardial
exist (119). An additional variation of this anomaly occurs ischemia are common presentations in symptomatic patients
when the LAD alone arises from the right sinus of Valsalva (85,134). Large fistulas are associated with the development of
or the proximal RCA. This is a very rare anomaly and several very tortuous ecstatic coronary arteries proximal to the origin
[2/7/010/8:36:2] [173–193]
Figure 16.18 Diagrams of the angiographic appearance of an anomalous left main coronary artery arising from the right sinus of Valsalva
shown in a 308 right anterior oblique projection. These illustrate the general appearance and relationship of the anomalous left main to the
aorta, the LAD and the LCx. In the anterior course (type A) the LCx is inferior to the left main and proximal LAD when there is no cranial
angulation used. With the interarterial course (type B), the left main may appear “on-end” as a dot as it courses around the aorta anteriorly and
the LCx arises with a caudal orientation. With the course via the crista supraventricularis or septum (type C), the LCx forms a cranial loop with
respect to the left main and the proximal left main gives rise to one or more septal vessels. Finally, the retroaortic, or dorsal (type D) course is
easily identified in the projection by the caudal and posterior initial course of the anomalous left main. Abbreviations: LAD, left anterior
descending; LCx, left circumflex. Source: Adapted from Ref. 111.
of the fistula. Fistulae can arise from any of the coronary In the normal adult heart, the collateral channels are small,
arteries, but about 50% arise from the RCA, 42% from the left thin-walled channels, usually less than 50 mm in diameter, that
coronary and 5% from both arteries (135). The most common contribute little to total coronary blood flow. However, in
sites for drainage of the fistulas are the coronary sinus, right response to coronary arterial narrowing and myocardial ische-
atrium or pulmonary artery, but many other locations have mia these collateral channels can increase in diameter to 200 to
been reported. Echocardiography with Doppler studies can 600 mm or greater, develop muscular media and transport a
suggest the presence of a fistula, but confirmation requires substantial proportion of blood flow (140). The enlargement of
computed tomographic or invasive angiographic imaging, the native channels is termed vasculogenesis, but collaterals can
latter method often coupled with quantification of the shunt also develop from the growth of new vessels (angiogenesis).
volume. If therapy if felt necessary, the treatment should oblit- Limited data exists regarding the conditions required to induce
erate the fistula yet maintain antegrade coronary flow (136). collateral growth in humans. Several clinical factors including
Traditional coronary artery bypass or coronary reimplantation coronary occlusion and ischemia, hypoxia and anemia have
with closure of the fistula have been used, but percutaneous been identified as stimulating collateral development and evi-
methods such as catheter embolization are now being consid- dence suggests this is mediated by the release of certain growth
ered as an alternative (137,138). Multiple coronary-cameral factors, cytokines and proteases from activated monocytes
fistulas causing myocardial ischemia have been reported in (141,142). Collateral development can occur quickly. At the
some patients (139). In these patients, coronary angiography time of acute coronary occlusion resulting in myocardial infarc-
shows diffuse endocardial opacification and filling of the tion, only 16% of patients had angiographically evident collat-
ventricular cavity. eral vessels. Serial angiography showed that effective
collaterals were present in 50% of the patients by two weeks
and 66% of patients by seven weeks following the infarction
CORONARY COLLATERAL BLOOD VESSELS (143). Functioning collaterals can maintain a coronary artery
Normal embryologic development of the coronary circulation wedge pressure that averages nearly 40% of the mean aortic
includes the formation of collateral vessels, which connect pressure thereby maintaining myocardial viability (144). The
different components of the coronary arterial circulation. angiographic demonstration of collateral flow to an area in the
[2/7/010/8:36:2] [173–193]
Figure 16.19 Diagram showing the course of an anomalous left main arising from the right sinus of Valsalva. When either the right or, as in
this illustration, left coronary artery arises from the opposite sinus of Valsalva and courses between the aorta and PA the anomalous artery
has an acute-angle takeoff and the normal round to oval coronary ostium is converted into a slit-like shape. Abbreviations: RCA, right
coronary artery; LCx, left circumflex artery; LAD, left anterior descending artery; PA, pulmonary artery. Source: Adapted from Ref. 112.
Figure 16.20 Large coronary-cameral fistula. Coro-

nary angiogram of the left coronary artery in a right
anterior oblique view (left). The origin of the fistula from
the left anterior descending artery is not clearly seen
and it appeared to drain into the coronary sinus. Multi-
slice coronary computed tomographic angiogram
showing the origin of the fistula from a dilated diagonal
branch and the small left anterior descending branch
beyond the origin of the diagonal (right). The fistula is
seen emptying into the left ventricular cavity. Source:
From Ref. 133, BMJ Publishing Group Ltd.
distribution of an occluded coronary artery is one of the artery or within the same branch via the vaso vasorum. Com-
strongest indicators on myocardial viability in the affected area. mon routes for collateral blood flow to the three main coronary
Functional collaterals can develop between the terminal arteries were elegantly defined by Levin in the early days of
extensions of two coronary arteries, between the side branches coronary angiography and are still relevant today (Fig. 16.21)
of the two coronary arteries, between branches of the same (145). Certain collaterals have been specifically identified such
[2/7/010/8:36:2] [173–193]
Figure 16.21 Common routes of collateral supply to

the three major coronary arteries. (A) Collateral
channels that develop with occlusion of the LAD.
(B) Collateral channels that develop with occlusion
of the RCA and (C) collateral channels that develop
with occlusion of the LCx. Abbreviations: LAD, left
anterior descending; RCA, right coronary artery; LCx,
left circumflex; AV, atrioventricular branch; D, diago-
nal branch; LAO, left anterior oblique; LCA, left cor-
onary artery; OM, obtuse marginal; PDA, posterior
descending artery; RAO, right anterior oblique.
[2/7/010/8:36:2] [173–193]
as Kugel’s artery (146). Although originally described as arising 19. Angelini P. Normal and anomalous coronary arteries: definitions
from the proximal LCx, most now consider a Kugel’s artery as a and classification. Am Heart J 1989; 117:418–434.
collateral arising from either the proximal RCA, the sinus node 20. Clark VL, Brymer JF, Lakier JB. Posterior descending artery
artery or LCx that traverses the intraatrial septum to anasto- origin from the left anterior descending: an unusual coronary
artery variant. Cathet Cardiovasc Diagn 1985; 11:167–171.
mose with the atrioventricular node artery thus supplying the
21. Musselman DR, Tate DA. Left coronary dominance due to direct
distal RCA or LCx branches. It is identified in 3% to 6% of continuation of the left anterior descending to form the posterior
angiograms in patients with significant coronary atherosclerosis descending coronary artery. Chest 1992; 102:319–320.
(147). Another specifically identified collateral is a Vieussens’ 22. Spindola-Franco H, Grose R, Solomon N. Dual left anterior
ring (148). This is a collateral connection between the conus descending coronary artery: angiographic description of impor-
artery of the RCA and a proximal right ventricular branch of the tant variants and surgical implications. Am Heart J 1983; 105:
LAD. This collateral circle provides flow to reconstitute a prox- 445–455.
imally occluded LAD or less frequently a proximally occluded 23. Yoshikai M, Hamada M, Murayama J, et al. A very rare anatomic
RCA. variation of the right ventricular branch: right ventricular
descending branch from the anterior interventricular branch.
Surg Radiol Anat 2004; 26:512–514.
24. Ilia R, Goldfarb B, Katz A, et al. Variations in blood supply to the
REFERENCES anterior interventricular septum: incidence and possible clinical
1. Furchgott RF, Zawadzki JV. The obligatory role of endothelial importance. Cathet Cardiovasc Diagn 1991; 24:277–282.
cells in the relaxation of arterial smooth muscle by acetylcholine. 25. Bream PR, Jones JM, Elliott LP. The anomalous septal perforating
Nature 1980; 288:373–376. artery. Its origin from the first diagonal, first marginal, or cir-
2. Moncada S, Vane JR. Arachidonic acid metabolites and the cumflex artery. Radiology 1981; 138:301–307.
interactions between platelets and blood-vessel walls. N Engl J 26. Rath S, Har-Zahav Y, Battler A, et al. Frequency and clinical
Med 1979; 300:1142–1147. significance of anomalous origin of septal perforator coronary
3. Yanagisawa M, Kurihara H, Kimura S, et al. A novel potent artery. Am J Cardiol 1986; 58:657–658.
vasoconstrictor peptide produced by vascular endothelial cells. 27. Nerantzis C, Avgoustakis D. An S-shaped atrial artery supplying
Nature 1988; 332:411–415. the sinus node area. An anatomical study. Chest 1980; 78:
4. Galis ZS, Muszynski M, Sukhova GK, et al. Cytokine-stimulated 274–278.
human vascular smooth muscle cells synthesize a complement of 28. Schlesinger MJ, Zoll PM, Wessler S. The conus artery; a third
enzymes required for extracellular matrix digestion. Circ Res coronary artery. Am Heart J 1949; 38:823–836, illust.
1994; 75:181–189. 29. Adams J, Treasure T. Variable anatomy of the right coronary
5. Sato T, Arai K, Ishiharajima S, et al. Role of glycosaminoglycan artery supply to the left ventricle. Thorax 1985; 40:618–620.
and fibronectin in endothelial cell growth. Exp Mol Pathol 1987; 30. Kyriakidis MK, Kourouklis CB, Papaioannou JT, et al. Sinus node
47:202–210. coronary arteries studied with angiography. Am J Cardiol 1983; 51:
6. Hannan RL, Kourembanas S, Flanders KC, et al. Endothelial cells 749–750.
synthesize basic fibroblast growth factor and transforming 31. Errichetti A, Mills RM Jr, Mercadante NM, et al. Anomalous
growth factor beta. Growth Factors 1988; 1:7–17. origin of the posterior descending artery from the first septal
7. Rosenberg RD, Rosenberg JS. Natural anticoagulant mechanisms. perforator. Cathet Cardiovasc Diagn 1986; 12:402–404.
J Clin Invest 1984; 74:1–6. 32. Voci P, Bilotta F, Caretta Q, et al. Papillary muscle perfusion
8. Clowes AW, Clowes MM, Fingerle J, et al. Regulation of smooth pattern. A hypothesis for ischemic papillary muscle dysfunction.
muscle cell growth in injured artery. J Cardiovasc Pharmacol Circulation 1995; 91:1714–1718.
1989; 14(suppl 6):S12–S15. 33. Kuo L, Chilian WM, Davis MJ. Coronary arteriolar myogenic
9. James TN. Anatomy of the Coronary Arteries. New York: P.B. response is independent of endothelium. Circ Res 1990; 66:
Hoeber, 1961. 860–866.
10. Kronzon I, Deutsch P, Glassman E. Length of the left main 34. Chilian WM. Coronary microcirculation in health and disease.
coronary artery: its relation to the pattern of coronary arterial Summary of an NHLBI workshop. Circulation 1997; 95:522–528.
distribution. Am J Cardiol 1974; 34:787–789. 35. Camici PG, Crea F. Coronary microvascular dysfunction. N Engl
11. Green GE, Bernstein S, Reppert EH. The length of the left main J Med 2007; 356:830–840.
coronary artery. Surgery 1967; 62:1021–1024. 36. von Lüdinghausen M. The venous drainage of the human
12. Virmani R, Chun PK, Robinowitz M, et al. Length of left main myocardium. Adv Anat Embryol Cell Biol 2003; 168:1–104.
coronary artery. Lack of correlation to coronary artery domi- 37. Murphy J. Applied anatomy of the heart and great vessels.
nance and bicuspid aortic valve: an autopsy study of 54 cases. In: Murphy J, ed. Mayo Clinic Cardiology Review. 2nd ed.
Arch Pathol Lab Med 1984; 108:638–641. Philadelphia: Lippincott Williams & Wilkins, 2005:948.
13. Bergelson BA, Tommaso CL. Left main coronary artery disease: 38. Hood WB Jr. Regional venous drainage of the human heart. Br
assessment, diagnosis, and therapy. Am Heart J 1995; 129:350– Heart J 1968; 30:105–109.
359. 39. Gerber TC, Sheedy PF, Bell MR, et al. Evaluation of the coronary
14. Levin DC, Harrington DP, Bettmann MA, et al. Anatomic varia- venous system using electron beam computed tomography. Int J
tions of the coronary arteries supplying the anterolateral aspect Cardiovasc Imaging 2001; 17:65–75.
of the left ventricle: possible explanation for the “Unexplained” 40. Gilard M, Mansourati J, Etienne Y, et al. Angiographic anatomy
anterior aneurysm. Invest Radiol 1982; 17:458–462. of the coronary sinus and its tributaries. Pacing Clin Electro-
15. Baptista CA, DiDio LJ, Prates JC. Types of division of the left physiol 1998; 21:2280–2284.
coronary artery and the ramus diagonalis of the human heart. 41. Bales GS. Great cardiac vein variations. Clin Anat 2004; 17:
Jpn Heart J 1991; 32:323–335. 436–443.
16. Schlesinger MJ. Relation of anastomotic pattern to pathologic 42. Mantini E, Grondin CM, Lillehei CW, et al. Congenital Anoma-
conditions of the coronary arteries. Arch Path 1940; 30:403–415. lies Involving the Coronary Sinus. Circulation 1966; 33:317–327.
17. Giuliani E, Gersh B, McGoon M, et al., eds. Mayo Clinic Practice 43. Ortale JR, Gabriel EA, Iost C, et al. The anatomy of the coronary
of Cardiology. 3rd ed. St. Louis, Missouri: Mosby, 1996:341. sinus and its tributaries. Surg Radiol Anat 2001; 23:15–21.
18. Allwork SP. The applied anatomy of the arterial blood supply to 44. Yamada T, McElderry HT, Plumb VJ, et al. Duplicated coronary
the heart in man. J Anat 1987; 153:1–16. sinus with a connecting branch. Europace 2008; 10:880–881.
[2/7/010/8:36:2] [173–193]
45. Shinbane JS, Girsky MJ, Mao S, et al. Thebesian valve imaging 67. Stolte M, Weis P, Prestele H. Muscle bridges over the left anterior
with electron beam CT angiography: implications for resynch- descending coronary artery: their influence on arterial disease
ronization therapy. Pacing Clin Electrophysiol 2004; 27: (author’s transl). Virchows Arch A Pathol Anat Histol 1977;
1566–1567. 375:23–36.
46. Singh JP, Houser S, Heist EK, et al. The coronary venous anat- 68. Angelini P, Trivellato M, Donis J, et al. Myocardial bridges: a
omy: a segmental approach to aid cardiac resynchronization review. Prog Cardiovasc Dis 1983; 26:75–88.
therapy. J Am Coll Cardiol 2005; 46:68–74. 69. Irvin RG. The angiographic prevalence of myocardial bridging in
47. Fishman MC, Chien KR. Fashioning the vertebrate heart: earliest man. Chest 1982; 81:198–202.
embryonic decisions. Development 1997; 124:2099–2117. 70. Roberts WC, Dicicco BS, Waller BF, et al. Origin of the left main
48. Mikawa T, Fischman DA. The polyclonal origin of myocyte from the right coronary artery or from the right aortic sinus with
lineages. Annu Rev Physiol 1996; 58:509–521. intramyocardial tunneling to the left side of the heart via the
49. Manasek FJ. Embryonic development of the heart. I. A light and ventricular septum. The case against clinical significance of
electron microscopic study of myocardial development in the myocardial bridge or coronary tunnel. Am Heart J 1982; 104:
early chick embryo. J Morphol 1968; 125:329–365. 303–305.
50. Reese DE, Mikawa T, Bader DM. Development of the coronary 71. Berry JF, von Mering GO, Schmalfuss C, et al. Systolic compres-
vessel system. Circ Res 2002; 91:761–768. sion of the left anterior descending coronary artery: a case series,
51. Levin DC, Beckmann CF, Garnic JD, et al. Frequency and clinical review of the literature, and therapeutic options including stent-
significance of failure to visualize the conus artery during coro- ing. Catheter Cardiovasc Interv 2002; 56:58–63.
nary arteriography. Circulation 1981; 63:833–837. 72. Chee TP, Jensen DP, Padnick MB, et al. Myocardial bridging of
52. Menke DM, Waller BF, Pless JE. Hypoplastic coronary arteries the left anterior descending coronary artery resulting in suben-
and high takeoff position of the right coronary ostium. A fatal docardial infarction. Arch Intern Med 1981; 141:1703–1704.
combination of congenital coronary artery anomalies in an ama- 73. Cutler D, Wallace JM. Myocardial bridging in a young patient
teur athlete. Chest 1985; 88:299–301. with sudden death. Clin Cardiol 1997; 20:581–583.
53. Roberts WC. Major anomalies of coronary arterial origin seen in 74. Dean JW, Mills PG. Abnormal ventricular repolarisation in asso-
adulthood. Am Heart J 1986; 111:941–963. ciation with myocardial bridging. Br Heart J 1994; 71:366–367.
54. Rath S, Battler A. Anomalous origin of the right coronary artery 75. Faruqui AM, Maloy WC, Felner JM, et al. Symptomatic
from the left anterior descending coronary artery. Cathet Car- myocardial bridging of coronary artery. Am J Cardiol 1978;
diovasc Diagn 1998; 44:328–329. 41:1305–1310.
55. Alexander RW, Griffith GC. Anomalies of the coronary arteries 76. Ge J, Erbel R, Rupprecht HJ, et al. Comparison of intravascular
and their clinical significance. Circulation 1956; 14:800–805. ultrasound and angiography in the assessment of myocardial
56. Blake HA, Manion WC, Mattingly TW, et al. Coronary Artery bridging. Circulation 1994; 89:1725–1732.
Anomalies. Circulation 1964; 30:927–940. 77. Klues HG, Schwarz ER, vom Dahl J, et al. Disturbed intracoro-
57. Dicicco BS, McManus BM, Waller BF, et al. Separate aortic ostium nary hemodynamics in myocardial bridging: early normalization
of the left anterior descending and left circumflex coronary by intracoronary stent placement. Circulation 1997; 96:2905–2913.
arteries from the left aortic sinus of Valsalva (absent left main 78. Mahon NG, Sugrue DD. Treatment of a long segment of symp-
coronary artery). Am Heart J 1982; 104:153–154. tomatic myocardial bridging with multiple coronary stents.
58. Arnett EN, Isner JM, Redwood DR, et al. Coronary artery J Invasive Cardiol 1997; 9:484–487.
narrowing in coronary heart disease: comparison of cineangio- 79. Vogt PR, Dubach P, Turina MI. Images in cardiovascular med-
graphic and necropsy findings. Ann Intern Med 1979; 91:350–356. icine. Muscle bridging of the left anterior descending coronary
59. Hutchins GM, Bulkley BH, Ridolfi RL, et al. Correlation of artery. Circulation 1996; 93:614.
coronary arteriograms and left ventriculograms with postmor- 80. Ishii T, Hosoda Y, Osaka T, et al. The significance of myocardial
tem studies. Circulation 1977; 56:32–37. bridge upon atherosclerosis in the left anterior descending cor-
60. Siegel RJ, Swan K, Edwalds G, et al. Limitations of postmortem onary artery. J Pathol 1986; 148:279–291.
assessment of human coronary artery size and luminal narrow- 81. Bilazarian SD, Jacobs AK, Fonger JD, et al. Case report of a
ing: differential effects of tissue fixation and processing on coronary anomaly: crossing obtuse marginal arteries. Cathet
vessels with different degrees of atherosclerosis. J Am Coll Cardiovasc Diagn 1991; 23:130–132.
Cardiol 1985; 5:342–346. 82. Muyldermans LL, Van den Heuvel PA, Ernst SM. Epicardial
61. Dodge JT Jr, Brown BG, Bolson EL, et al. Lumen diameter of crossing of coronary arteries: a variation of coronary arterial
normal human coronary arteries. Influence of age, sex, anatomic anatomy. Int J Cardiol 1985; 7:416–419.
variation, and left ventricular hypertrophy or dilation. Circula- 83. Carangal VP, Dehmer GJ. Intercoronary communication between
tion 1992; 86:232–246. the circumflex and right coronary arteries. Clin Cardiol 2000;
62. Vieweg WV, Alpert JS, Hagan AD. Caliber and distribution of 23:125–126.
normal coronary arterial anatomy. Cathet Cardiovasc Diagn 84. Donaldson RF, Isner JM. Intercoronary continuity: an anatomic
1976; 2:269–280. basis for bidirectional coronary blood flow distinct from coro-
63. Sheifer SE, Canos MR, Weinfurt KP, et al. Sex differences in nary collaterals. Am J Cardiol 1984; 53:351–352.
coronary artery size assessed by intravascular ultrasound. Am 85. Sapin P, Frantz E, Jain A, et al. Coronary artery fistula: an
Heart J 2000; 139:649–653. abnormality affecting all age groups. Medicine 1990; 69:101–113.
64. Seiler C, Kirkeeide RL, Gould KL. Basic structure-function rela- 86. Hobbs RE, Millit HD, Raghavan PV, et al. Coronary artery
tions of the epicardial coronary vascular tree. Basis of quantita- fistulae: a 10-year review. Cleve Clin Q 1982; 49:191–197.
tive coronary arteriography for diffuse coronary artery disease. 87. Baltaxe HA, Wixson D. The incidence of congenital anomalies of
Circulation 1992; 85:1987–2003. the coronary arteries in the adult population. Radiology 1977; 122:
65. Ayala F, Badui E, Murillo H, et al. Right coronary ostium 47–52.
agenesis with anomalous origin of the right coronary artery 88. Chaitman BR, Lesperance J, Saltiel J, et al. Clinical, angiographic,
from an ectasic circumflex artery. A case report. Angiology and hemodynamic findings in patients with anomalous origin of
1995; 46:637–639. the coronary arteries. Circulation 1976; 53:122–131.
66. Kramer JR, Kitazume H, Proudfit WL, et al. Clinical significance 89. Cieslinski G, Rapprich B, Kober G. Coronary anomalies: inci-
of isolated coronary bridges: benign and frequent condition dence and importance. Clin Cardiol 1993; 16:711–715.
involving the left anterior descending artery. Am Heart J 1982; 90. Engel HJ, Torres C, Page HL Jr. Major variations in anatomical
103:283–288. origin of the coronary arteries: angiographic observations in
[2/7/010/8:36:2] [173–193]
4,250 patients without associated congenital heart disease. Cathet 112. Kragel AH, Roberts WC. Anomalous origin of either the right or
Cardiovasc Diagn 1975; 1:157–169. left main coronary artery from the aorta with subsequent cours-
91. Kimbiris D, Iskandrian AS, Segal BL, et al. Anomalous aortic ing between aorta and pulmonary trunk: analysis of 32 necropsy
origin of coronary arteries. Circulation 1978; 58:606–615. cases. Am J Cardiol 1988; 62:771–777.
92. Topaz O, DeMarchena EJ, Perin E, et al. Anomalous coronary 113. Taylor AJ, Rogan KM, Virmani R. Sudden cardiac death asso-
arteries: angiographic findings in 80 patients. Int J Cardiol 1992; ciated with isolated congenital coronary artery anomalies. J Am
34:129–138. Coll Cardiol 1992; 20:640–647.
93. Wilkins CE, Betancourt B, Mathur VS, et al. Coronary artery 114. Barth CW III, Roberts WC. Left main coronary artery originating
anomalies: a review of more than 10,000 patients from the Clayton from the right sinus of Valsalva and coursing between the aorta
Cardiovascular Laboratories. Tex Heart Inst J 1988; 15:166–173. and pulmonary trunk. J Am Coll Cardiol 1986; 7:366–373.
94. Yamanaka O, Hobbs RE. Coronary artery anomalies in 126,595 115. Cheitlin MD, De Castro CM, McAllister HA. Sudden death as a
patients undergoing coronary arteriography. Cathet Cardiovasc complication of anomalous left coronary origin from the anterior
Diagn 1990; 21:28–40. sinus of Valsalva, A not-so-minor congenital anomaly. Circula-
95. Hanley FL, Sade RM, Blackstone EH, et al. Outcomes in neonatal tion 1974; 50:780–787.
pulmonary atresia with intact ventricular septum. A multiinstitu- 116. Roberts WC, Kragel AH. Anomalous origin of either the right or
tional study. J Thorac Cardiovasc Surg 1993; 105:406–423, 424–427; left main coronary artery from the aorta without coursing of the
discussion 423–424. anomalistically arising artery between aorta and pulmonary
96. Sim EK, van Son JA, Edwards WD, et al. Coronary artery anat- trunk. Am J Cardiol 1988; 62:1263–1267.
omy in complete transposition of the great arteries. Ann Thorac 117. Isner JM, Shen EM, Martin ET, et al. Sudden unexpected death as
Surg 1994; 57:890–894. a result of anomalous origin of the right coronary artery from the
97. White NK, Edwards JE. Anomalies of the coronary arteries; left sinus of Valsalva. Am J Med 1984; 76:155–158.
report of four cases. Arch Pathol (Chic) 1948; 45:766–771. 118. Roberts WC, Siegel RJ, Zipes DP. Origin of the right coronary artery
98. Page HL Jr, Engel HJ, Campbell WB, et al. Anomalous origin of from the left sinus of valsalva and its functional consequences:
the left circumflex coronary artery. Recognition, antiographic analysis of 10 necropsy patients. Am J Cardiol 1982; 49:863–868.
demonstration and clinical significance. Circulation 1974; 119. Brothers JA, Stephens P, Gaynor JW, et al. Anomalous aortic
50:768–773. origin of a coronary artery with an interarterial course: should
99. Piovesana P, Corrado D, Verlato R, et al. Morbidity associated family screening be routine? J Am Coll Cardiol 2008; 51:2062–2064.
with anomalous origin of the left circumflex coronary artery from 120. Coyle L, Thomas WJ. Anomalous left anterior descending coro-
the right aortic sinus. Am J Cardiol 1989; 63:762–763. nary artery: malignant hospital course of a not so benign anom-
100. Rozenman Y, Schechter D, Gilon D, et al. Anomalous origin of aly. Catheter Cardiovasc Interv 2000; 51:468–470.
the circumflex coronary artery from the right sinus of Valsalva as 121. Russo G, Tamburino C, Licciardello G, et al. Isolated, anomalous
a cause of ischemia at old age. Clin Cardiol 1993; 16:900–901. origin of the left anterior descending coronary artery from the
101. Click RL, Holmes DR Jr, Vlietstra RE, et al. Anomalous coronary right coronary artery with angina pectoris. Eur Heart J 1991;
arteries: location, degree of atherosclerosis and effect on 12:558–560.
survival–a report from the Coronary Artery Surgery Study. J 122. Tuncer C, Batyraliev T, Yilmaz R, et al. Origin and distribution
Am Coll Cardiol 1989; 13:531–537. anomalies of the left anterior descending artery in 70,850 adult
102. Roberts WC, Morrow AG. Compression of anomalous left cir- patients: multicenter data collection. Catheter Cardiovasc Interv
cumflex coronary arteries by prosthetic valve fixation rings. 2006; 68:574–585.
J Thorac Cardiovasc Surg 1969; 57:834–838. 123. Dogan SM, Gursurer M, Aydin M, et al. Myocardial ischemia
103. Lipton MJ, Barry WH, Obrez I, et al. Isolated single coronary caused by a coronary anomaly left anterior descending coronary
artery: diagnosis, angiographic classification, and clinical signif- artery arising from right sinus of Valsalva. Int J Cardiol 2006; 112:
icance. Radiology 1979; 130:39–47. e57–e59.
104. Ogden JA, Goodyer AV. Patterns of distribution of the single 124. Roynard JL, Cattan S, Artigou JY, et al. Anomalous course of the
coronary artery. Yale J Biol Med 1970; 43:11–21. left anterior descending coronary artery between the aorta and
105. Wesselhoeft H, Fawcett JS, Johnson AL. Anomalous origin pulmonary trunk: a rare cause of myocardial ischaemia at rest.
of the left coronary artery from the pulmonary trunk. Its clinical Br Heart J 1994; 72:397–399.
spectrum, pathology, and pathophysiology, based on a review 125. Takenaka T, Horimoto M, Fujiwara M. Anomalous origin of the
of 140 cases with seven further cases. Circulation 1968; 38: left anterior descending coronary artery from the right sinus of
403–425. Valsalva associated with effort angina pectoris. Eur Heart J 1993;
106. Frommelt PC, Frommelt MA. Congenital coronary artery anoma- 14:29–31.
lies. Pediatr Clin North Am 2004; 51:1273–1288. 126. Shirani J, Roberts WC. Solitary coronary ostium in the aorta in
107. Bland E, White P, Garland J. Congenital anomalies of the coro- the absence of other major congenital cardiovascular anomalies.
nary arteries: report of an unusual case associated with cardiac J Am Coll Cardiol 1993; 21:137–143.
hypertrophy. Am Heart J 1933; 8:787–801. 127. Basso C, Maron BJ, Corrado D, et al. Clinical profile of congenital
108. Thomas CS Jr, Campbell WB, Alford WC Jr, et al. Complete coronary artery anomalies with origin from the wrong aortic
repair of anomalous origin of the left coronary artery in the adult. sinus leading to sudden death in young competitive athletes.
J Thorac Cardiovasc Surg 1973; 66:439–446. J Am Coll Cardiol 2000; 35:1493–1501.
109. Michielon G, Di Carlo D, Brancaccio G, et al. Anomalous coro- 128. Angelini P, Flamm SD. Newer concepts for imaging anomalous
nary artery origin from the pulmonary artery: correlation aortic origin of the coronary arteries in adults. Catheter Cardio-
between surgical timing and left ventricular function recovery. vasc Interv 2007; 69:942–954.
Ann Thorac Surg 2003; 76:581–588. 129. Deibler AR, Kuzo RS, Vohringer M, et al. Imaging of congenital
110. Kandzari DE, Harrison JK, Behar VS. An anomalous left coronary coronary anomalies with multislice computed tomography.
artery originating from the pulmonary artery in a 72-year-old Mayo Clin Proc 2004; 79:1017–1023.
woman: diagnosis by color flow myocardial blush and coronary 130. Byrum CJ, Blackman MS, Schneider B, et al. Congenital atresia of
arteriography. J Invasive Cardiol 2002; 14:96–99. the left coronary ostium and hypoplasia of the left main coronary
111. Ishikawa T, Brandt PW. Anomalous origin of the left main artery. Am Heart J 1980; 99:354–358.
coronary artery from the right anterior aortic sinus: angiographic 131. Ueda K, Saito A, Nakano H, et al. Absence of proximal coronary
definition of anomalous course. Am J Cardiol 1985; 55: arteries associated with pulmonary atresia. Am Heart J 1983;
770–776. 106:596–598.
[2/7/010/8:36:2] [173–193]
132. Musiani A, Cernigliaro C, Sansa M, et al. Left main coronary 141. Lazarous DF, Scheinowitz M, Shou M, et al. Effects of chronic
artery atresia: literature review and therapeutical considerations. systemic administration of basic fibroblast growth factor on
Eur J Cardiothorac Surg 1997; 11:505–514. collateral development in the canine heart. Circulation 1995;
133. Ho HH, Cheung CW, Jim MH, et al. Images in cardiology: 91:145–153.
coronary-cameral fistula. Heart 2005; 91:1540. 142. Zoll PM, Wessler S, Schlesinger MJ. Interarterial coronary anas-
134. Luo L, Kebede S, Wu S, et al. Coronary artery fistulae. Am J Med tomoses in the human heart, with particular reference to anemia
Sci 2006; 332:79–84. and relative cardiac anoxia. Circulation 1951; 4:797–815.
135. Levin DC, Fellows KE, Abrams HL. Hemodynamically signifi- 143. Schwartz H, Leiboff RH, Bren GB, et al. Temporal evolution of
cant primary anomalies of the coronary arteries. Angiographic the human coronary collateral circulation after myocardial
aspects. Circulation 1978; 58:25–34. infarction. J Am Coll Cardiol 1984; 4:1088–1093.
136. Rittenhouse EA, Doty DB, Ehrenhaft JL. Congenital coronary 144. Seiler C, Fleisch M, Garachemani A, et al. Coronary collateral
artery-cardiac chamber fistula. Review of operative management. quantitation in patients with coronary artery disease using intra-
Ann Thorac Surg 1975; 20:468–485. vascular flow velocity or pressure measurements. J Am Coll
137. Perry SB, Rome J, Keane JF, et al. Transcatheter closure of Cardiol 1998; 32:1272–1279.
coronary artery fistulas. J Am Coll Cardiol 1992; 20:205–209. 145. Levin DC. Pathways and functional significance of the coronary
138. Urrutia SC, Falaschi G, Ott DA, et al. Surgical management of collateral circulation. Circulation 1974; 50:831–837.
56 patients with congenital coronary artery fistulas. Ann Thorac 146. Kugel M. Anatomical studies on the coronary arteries and their
Surg 1983; 35:300–307. branches. I. Arteria anastomotica auricularis magna. Am Heart J
139. Wolf A, Rockson SG. Myocardial ischemia and infarction due to 1927; 3:260–270.
multiple coronary-cameral fistulae: two case reports and review 147. Grollman JH Jr, Heger L. Angiographic anatomy of the left
of the literature. Cathet Cardiovasc Diagn 1998; 43:179–183. Kugel’s artery. Cathet Cardiovasc Diagn 1978; 4:127–133.
140. Gregg DE, Patterson RE. Functional importance of the coronary 148. O’Leary EL, Garza L, Williams M, et al. Images in cardiovascular
collaterals. N Engl J Med 1980; 303:1404–1406. medicine. Vieussens’ ring. Circulation 1998; 98:487–488.
[15/7/010/22:31:46] [194–207]
17
Diagnostic angiographic catheters: coronary and vascular

Ryan Berg and Michael Lim
INTRODUCTION of the selected vessel, the smaller catheters are intended to be

While hundreds of catheters are available for use in the less traumatic to the selected vessel and allow for smaller
catheterization laboratory, most operators have a select few percutaneous entry sheaths. Standard selective peripheral arte-
that compromise the vast majority of use. In this chapter, we rial catheters are similar in construction but they come in smaller
provide a practical guide for many of the most important lengths as well. Most diagnostic catheters are rated to accept a
“workhorse” catheters in general use today. We do not intend maximum of 1200 psi of contrast injection.
to provide an exhaustive list of all available catheters; rather,
we discuss the principal catheters used in the fields of coronary
angiography, ventriculography, and peripheral angiography. LEFT CORONARY CATHETERS
Over two-thirds of the mortality risk associated with diagnostic
coronary catheterization is related to trauma of the left main
HISTORY OF SELECTIVE ANGIOGRAPHIC coronary artery (1). Therefore, proper diagnostic catheter selec-
CATHETERS tion and proper engagement of the left main is of critical
Dr Mason Sones performed the first diagnostic coronary angio- importance. The most common diagnostic coronary catheter
gram in 1958. He later developed a standard technique from a used to engage the ostium of the left main is the Judkins left.
brachial cutdown approach with design of his own catheters. This endhole catheter has a 908 primary curve to allow the tip to
Sones helped develop a single catheter with a stiff body to enter the coronary ostium and a 1808 secondary curve to allow
allow for torqueability and a tapered tip that allowed it to be backup support from the opposing wall of the aorta. Depend-
manipulated against the aortic cusps to take a primary curve to ing on the manufacturer, it might go by a different brand name
engage the coronary ostia. The tapered tip also prevented the or abbreviation, but it is usually abbreviated “JL” for Judkins
tip of the catheter from completely obstructing the coronary left. The length of the segment between the primary and
ostia. This is known as the Sones catheter and the technique for secondary curve is denoted by a number, so JL4 means there
engaging the coronary ostia through the brachial cutdown is is 4 cm in length between the primary and secondary curves.
known as the Sones technique. The Judkins left catheter comes typically in sizes ranging from
The next major milestone in coronary catheter design 3.5 to 6 cm (Fig. 17.2). Bigger sizes are needed for larger patients
came from Dr Melvin Judkins, who developed preshaped and those with dilatation of the aortic arch.
selective coronary catheters based on chest radiographs. The JL catheters are the most commonly used because
These catheters are now known as the Judkins left and right very little manipulation is needed to engage most left main
catheters, and his technique of inserting them from a percuta- arteries. Generally, the catheter is advanced under fluoroscopy
neous femoral approach is known as the Judkins technique. to the ascending aorta over a wire. After the wire is removed,
While the Sones technique is of tremendous historical impor- the catheter tends to move cranially and leftward to engage the
tance, most present-day operators have not been trained in his left main artery. The left main can be engaged with usually only
brachial cutdown approach with Sones catheters. A percuta- a slight advancement of the catheter forward at the introducer
neous approach and standardized preformed catheters are now sheath, seen as a “pop” of the catheter into the left main while
the standard from both femoral and arm (radial and brachial) watching on fluoroscopy. Rarely, the left main can be out of
access for both coronary and peripheral angiography. plane either anteriorly or posteriorly and then either slight
clockwise or counterclockwise torque (respectively) may be
used to engage the artery. Another common problem is choos-
GENERAL DIAGNOSTIC CATHETER DESIGN ing the wrong initial catheter size. Rarely, a Judkins left of any
Modern vascular diagnostic catheters are typically polymer- size is not able to properly engage the left main satisfactorily for
blended catheters made of nylon, polyurethane, or other pro- adequate angiography. This situation can occur when the left
prietary polymers. Polymer catheters are reinforced with main orifice is too high or too out of plane, or when the left
braided steel that promotes torque transmission from the distal main is very short and selective injection of the left anterior
end to the tip. The catheter tip is usually tapered with a non- descending and left circumflex is needed. In these cases, a more
braided polymer that allows for increased flexibility and less torqueable catheter such as the Amplatz left, invented by Dr
trauma when being manipulated into a vessel ostium (Fig. 17.1). Kurt Amplatz, can be used.
Some catheters incorporate a radiopaque tip for maximal visu- The most common sizes of the Amplatz left diagnostic
alization. The standard selective diagnostic coronary catheters catheters are the AL1, AL2, and AL3. They are sized on the
are 100 cm in length and range in usual diameters from 4 to basis of the diameter of their secondary curve (Fig. 17.3). The
6 French (Fr). While larger sizes can deliver larger volumes and larger Amplatz catheters have more “reach” to engage more
increased flow rates of contrast allowing for better visualization superior takeoffs of coronary ostia. Typically the AL2 is used as
[15/7/010/22:31:46] [194–207]
DIAGNOSTIC ANGIOGRAPHIC CATHETERS: CORONARY AND VASCULAR 195
Figure 17.1 Catheters are created out of polymer and reinforced

with braided steel that allows torque transmission. The tip of the
catheter is usually tapered with a nonbraided polymer, which allows
for increased flexibility and less trauma.
Figure 17.4 Technique for cannulating the coronary arteries utiliz-

ing the Amplatz left catheter. The catheter is advanced until the
secondary curve rests in the noncoronary or posterior aortic cusp
and its tip is pointing toward the left coronary artery (A). The
catheter is then gently advanced and retracted until the ostium of
the coronary is engaged (B). Source: Adapted from Ref. 2.
engaging the left main. Amplatz left catheters are inserted into
the ascending aorta in standard fashion over a wire. When the
wire is removed, often the catheter will be positioned with its
tip pointing down to the aortic valve. If the tip is tilted toward
the right cusp, the catheter must be rotated so that the tip looks
toward the left cusp. The catheter must then be advanced so
that the secondary curve is sitting on the aortic valve with the
Figure 17.2 Judkins left coronary catheters in common sizes. tip climbing up the aortic wall and pointing slightly upward.
At this point, the catheter can be torqued either clockwise or
counterclockwise (tip anterior or posterior) to engage the
ostium of the left main (Fig. 17.4). There are various methods
to disengage an Amplatz left diagnostic catheter. Commonly,
once the Amplatz left catheter is engaged in the left main,
pushing the catheter will advance the front of the secondary
curve against the aortic wall thereby backing out the tip of the
catheter. In contrast, pulling the catheter may force the tip
deeper into the left main, increasing the risk of catheter dissec-
tion. Depending on how the secondary curve of the catheter is
sitting on the aortic root or how deeply the tip is engaged, this
“withdrawal paradox” may not occur.
RIGHT CORONARY CATHETERS

The most commonly used diagnostic right coronary catheter is
the Judkins right. Its primary curve is a 908 bend to allow the
catheter to enter the right coronary ostium. The secondary
curve is a gradual 308 bend. As with the Judkins left, depending
Figure 17.3 Amplatz left catheters in typical sizes reflective of the on the manufacturer, it might go by a different brand name or
diameter of their secondary curve. abbreviation, but it is usually abbreviated as “JR” for Judkins
right, followed by a number representing the length of the
distance between the primary and secondary curves. For most
the standard size to engage the left main coronary ostia, but typical adults, a JR4 catheter is used to engage the right coro-
higher or more superior directed left main ostia require longer nary artery. The Judkins right comes in sizes ranging from 3.5
catheters. The Amplatz left catheters require slightly more to 6 cm (Fig. 17.5). Bigger sizes are needed for patients with
manipulation, with subsequent increased risk for coronary larger or dilated ascending aortas. Unlike the Judkins left, the
dissection for engagement and disengagement than the stan- Judkins right catheter requires significant more manipulation
dard Judkins left catheters, making them a secondary choice for to engage the coronary ostium. The standard approach involves
[15/7/010/22:31:46] [194–207]
Figure 17.5 Judkins right catheters in typically available sizes.
inserting the catheter so that it rests on the cusps of the aortic

valve. The tip of the catheter starts facing toward the right of
the viewing monitor (toward the patient’s left). The catheter is
then pulled up to the level where it is assumed the right
coronary comes off (usually a centimeter or two above the
right coronary cusp), while torque in a clockwise direction is
simultaneously applied to rotate the tip so it slowly turns
toward the left of the screen and engages the ostium of the
right coronary artery. Sometimes it will not be possible to
engage from the level of the ostium, as the ostium lies just
below the sinotubular junction. The sinotubular junction of the
aorta in some patients is more pronounced and acts as a ridge Figure 17.6 Typical variations that may be found in the origination
that deflects the catheter tip away from the ostium. An alter- and proximal course of the right coronary artery. Source: From Ref. 3.
native approach is to start with the catheter slightly above the
level of the right coronary ostium, and torque clockwise while
simultaneously advancing downward toward the aortic valve.
In some patients, the JR4 catheter might not torque ade-
quately with clockwise rotation. In this circumstance, the cath- engagement. For a more superior angled takeoff, a hockey stick
eter is first torqued counterclockwise to see if that resolves the catheter or the El Gamal catheter (longer tip than the hockey
problem. It is important to avoid vigorous torqueing, as this can stick) can be used. Its primary curve can vary, but typically is
lead to kinking of the catheter. This will be recognized by a 758, allowing engagement of a more superior takeoff.
sudden dampening of the arterial pressure waveform, despite For a more anterior takeoff of the right coronary ostium,
little movement of the catheter itself. If the JR4 will not and initial catheter choice may be a Williams right, as described
cannulate the right coronary artery, the operator must consider previously. If the takeoff has a more significant anterior and
looking for the origin of the vessel in an anterior, posterior, or superior takeoff, an Amplatz right modified catheter can be
superior position and/or consider changing to an alternative used. The Amplatz right resembles the duckbill shape of the
catheter. A typical second-choice catheter for use in right cor- Amplatz left catheter, but it has a much smaller secondary
onary artery cannulation is the “no-torque” right catheter, also curve diameter. If more reach is needed for an even higher and
known as a Williams right or a 3DRC catheter. This catheter more anterior takeoff of the right coronary artery, an Amplatz
was designed to engage the right coronary ostium without any left catheter will be needed.
significant manipulation. It has a third curve out of plane of the
primary and secondary curves of the usual JR4 that gives it a
three-dimensional shape and allows it to engage the right BYPASS GRAFT CATHETERS
coronary ostium without significant torque. Common grafts that need to be engaged include the left IM
In some patients, the right coronary ostium has an alter- artery, the right internal mammary artery (RIMA), free radial or
native anatomic takeoff compared with the usual orthogonal saphenous vein grafts off of the anterior aorta that can course to
takeoff at the level of the free margin of the aortic cusp (3). any of the major epicardial vessels or rarely a gastroepiploic
Common variants include a superior takeoff that mimics the artery.
shape of the head of a shepherd’s crook or orthogonal takeoffs Typically the JR4 catheter is utilized as the workhorse in
from a more anterior position of the aortic wall (Fig. 17.6). In angiography of grafts. While the JR4 can commonly engage
the case of a shepherd’s crook origin, a catheter that has a tip most of the free grafts off of the anterior aorta going to the left-
that looks superior is necessary. One example is an internal sided epicardial vessels, it may not provide engagement of the
mammary (IM) artery catheter. The IM catheter is a modified usually inferior directed takeoff of a graft to the right coronary
JR4 with an 808 primary curve and a longer distal tip. This 108 artery. Also, the angle of the tip is sometimes not acute enough
difference in angle from the JR4 allows for a slight superior to selectively engage the ostium of the left and right internal
[15/7/010/22:31:46] [194–207]
while the catheter is then pulled back slightly to engage the

takeoff of the right subclavian artery. It is important to adjust the
torque of the catheter as it is being pulled back to always keep
the tip pointed directly cranial. The tip will otherwise have a
natural tendency to torque out of plane and the result is that the
catheter will be pulled back into the descending aorta without
engagement of the arch vessels. Once the catheter is seen to
jump into the brachiocephalic trunk, a test injection is given to
confirm the location and to get an idea of the takeoff of the
RIMA. The catheter tip can be manipulated to point more
toward the right subclavian compared with the right carotid.
A J-wire is then placed deep into the right subclavian and the
JR4 can be advanced to the mid portion of the right subclavian,
at least 2 cm distal to where the suspected takeoff of the RIMA is
Figure 17.7 Catheters frequently used for cannulation of bypass located. With frequent test injections, the JR4 is slowly pulled
grafts. back and torqued as needed to engage the RIMA. If the angle of
takeoff is steeper than can be engaged by the JR4, an exchange
length guide wire is needed to exchange the JR4 for the IM
diagnostic catheter. The 808 primary curve on the tip of the IM
thoracic arteries. In cases where selective engagement of the catheter as compared with the 908 tip of the JR4 should allow
grafts cannot be accomplished with a JR4, a specialty diagnostic successful selective engagement (Fig. 17.7).
catheter must be used. In a manner similar to engagement of the brachiocephalic
Free grafts to the distal right coronary artery or posterior trunk, the JR4 can be used to engage the left subclavian artery.
descending artery tend to have an inferior takeoff coming from Some operators prefer to engage the brachiocephalic trunk as
the right and anterior wall of the ascending aorta. Most described above, and then while keeping cranial angulation of
frequently, a multipurpose endhole catheter is the first choice the catheter tip, the catheter is pulled back until it is seen to jump
for engaging this graft. It is advanced over a wire with the into the left carotid artery, and then pulled back again until a
image intensifier in the left anterior oblique 308 position. The second jump is seen into the left subclavian. Alternatively, some
catheter begins with its tip high in the ascending aorta and is operators prefer to avoid engagement of the carotid artery in any
then turned so the tip faces the aortic wall to the left of the capacity, so the JR4 is brought with tip oriented caudally around
viewing screen (right of the patient). The tip of the catheter then to the proximal portion of the aortic arch and from there is
slowly is advanced downward while applying torque until it is pointed cranially to directly engage the left subclavian. With
seen or felt to fall into the right-sided graft. The multipurpose either technique, a test injection is done to confirm that the
catheter tends to work best in grafts that come off with a very catheter is in the left subclavian and to assess the general posi-
significant downward takeoff. For more subtle inferior takeoffs, tion of the takeoff of the left IM artery. A J-wire is then advanced
an Amplatz right modified catheter or a right coronary bypass into the left subclavian artery and the JR4 can be placed in the
(RCB) catheter can be helpful (Fig. 17.7). The RCB resembles a mid portion of the left subclavian, at least 1 to 2 cm distal to the
JR4 with a tip curved >908 so when it comes up and around the suspected takeoff of the LIMA. At this point the image intensifier
aorta, the tip of the catheter points slightly inferiorly. can be moved to the PA position if preferred. With frequent test
For grafts to the left anterior descending or left circumflex injections, the JR4 is slowly pulled back and torqued as needed
arteries or their branches, a JR4 catheter is the first choice and to engage the left IM artery. If the angle of takeoff is steeper than
will typically engage selectively. These grafts come off above can be engaged by the JR4, an exchange length guide wire is
the left main ostium with the left anterior descending grafts needed to exchange for the IM diagnostic catheter.
typically in the lowest position followed by the diagonal grafts Uncommonly, the right gastroepiploic artery might be
and then left circumflex grafts most superior. In cases where used as a bypass graft to the posterior descending artery (5).
the grafts are coming off with a more upward takeoff, the The right gastroepiploic artery originates as a branch of the
second choice is a left coronary bypass (LCB) catheter. The gastroduodenal artery, which is a branch of the common
primary curve is 908 like a JR4, but the secondary curve is 708 hepatic artery from the celiac trunk. Selective engagement of
(as compared with the gentle 308 secondary curve of the JR4), the celiac trunk will provide “nonselective” angiography of the
which gives the catheter more reach to the left side of the aorta right gastroduodenal artery. The celiac trunk originates from
as well as giving a slightly superior oriented tip of the catheter the anterior aorta just below the diaphragm at the level of the
(Fig.17.8). For grafts that cannot be engaged with a JR4 or LCB T12 vertebral body. Lateral angiography provides the best view
catheter, manipulation of an AL2, AL3, or multipurpose cath- of the takeoff of the celiac trunk to allow for selective engage-
eter frequently achieves the ability to selectively engage and ment with a JR4 or IMA catheter. When those catheters will not
image the vessel. selectively engage the celiac trunk, the visceral catheters such
For engaging the IM arteries, again a JR4 can be tried for as the Cobra or Sos (see section on peripheral diagnostic
initial engagement, as it will often selectively engage both the catheters) can be used. If poor visualization of the distal graft
right and left IM artery and can save a step of catheter exchange. anatomy and runoff is seen with nonselective angiography,
To manipulate the JR4 to the right subclavian, the catheter is super-selective angiography must be performed. This requires
pulled back to the top of the ascending aorta in the LAO 308 using an exchange length angled Terumo glide wire to advance
view, which lays out the aorta without foreshortening of the selectively into the right gastroepiploic (Fig. 17.8). The support-
branch vessels. In this view the tip of the JR4 is usually pointed ing diagnostic catheter can then be exchanged for a 4-Fr soft
caudally and is then rotated 1808 to point in a cranial direction glide catheter (Terumo) that is small and soft enough to be
[15/7/010/22:31:46] [194–207]
Figure 17.9 Commonly available multipurpose catheters.
Figure 17.8 Selective engagement of the GEA. First, a diagnostic

catheter is used to engage the CT. An exchange length angled glide
wire is then used to advance to the distal GEA graft. From here (not
pictured), the original diagnostic catheter can be walked out and
exchanged for a long glide catheter. The other branches of the CT
depicted are the SA, CHA, GDA, PHA, and the PD. Abbreviations:
CT, celiac trunk; GEA, gastroepiploic artery; SA, splenic artery;
CHA, common hepatic artery; GDA, gastroduodenal artery; PHA,
proper hepatic artery; PD, pancreatoduodenal artery; PDA, posterior
descending artery. Source: From Ref. 4.
positioned deep into the right gastroepiploic without causing

trauma. The 4-Fr size will allow for adequate angiography
when it is selectively engaged.
MULTIPURPOSE CATHETER
As previously mentioned, the multipurpose catheter is similar
in shape to the original Sones catheter with a straight tip and
Figure 17.10 (A) Technique of multipurpose catheter manipulation
only one curve. It was developed by Spencer King and Fred
for cannulation of the left coronary artery in the 308 RAO projection.
Schoonmaker (6). There are four types of common multipurpose
The tip of the catheter is slowly advanced off of the left coronary
catheters: the MPA1 and MPA2, and the MPB1 and MPB2 cusp with simultaneous counterclockwise rotation until in engages
(Fig. 17.9). The “A” curved catheters have approximately a the left coronary ostium. (B) Catheter manipulation for engagement
1208 curve while the B catheters have a 908 gradual curve. The of the right coronary artery ostium from the 608 LAO projection.
“1” designation refers to endhole only catheters and the “2” Abbreviation: RCA, right coronary artery. Source: From Ref. 6.
designation refers to the addition to two sideholes to the end-
hole. The sideholes allow for ventriculography on top of selec-
tive coronary angiography. Similar to the original Sones
technique, an operator can complete an entire diagnostic car- engage the coronaries, and care must be taken to avoid this
diac catheterization with just the multipurpose catheter. The predilection. With sideholes, a multipurpose catheter can be
soft tip allows it to be easily manipulated into a “J” shape when manipulated into the left ventricle and left ventricular angiog-
pressed against the aortic valve, thereby allowing it to more raphy can also be performed. Since the invention of more
easily selectively cannulate the left and right coronary ostium selective catheters as described above, this technique has
(Fig. 17.10). Because the tip is long, it can tend to more deeply fallen out of favor. Most common in today’s practice, the
[15/7/010/22:31:46] [194–207]
multipurpose catheter is used as a first or second choice for a

RCB graft angiography, or when multiple catheters have been
tried and failed to engage a native coronary or bypass graft, the
multipurpose can usually be manipulated into selective engage-
ment with trial and error.
CATHETERS FOR RADIAL OR BRACHIAL

APPROACH
Because of anatomy or patient preference, a radial or brachial
approach might be used for access. The right radial or brachial
approach is usually preferred as it allows the operator to stand
in the usual position on the right side of the normally posi-
tioned patient. Also, approaches from the right arm allow
direct access into the ascending aorta, whereas left arm
approaches end in the takeoff of the left subclavian artery,
which will frequently direct wires and catheters down the
descending aorta preferentially. Depending on which arm is
used for access, different types and sizes of catheters can be
used. In general, since the left arm approach takes the same
approach across the aortic arch and down the ascending aorta
as a femoral approach, the typical coronary catheters described
above can be used. Typically, a JL4 and JR4 can successfully be
used for coronary angiography from a left arm access.
From a right arm access, because of the straight trajectory
into the ascending aorta, half-size smaller Judkins left catheters Figure 17.11 Angulation and design of the Kimny curved catheter.
must be used than would normally be used for a particular
patient (e.g., start with a JL 3.5 instead of a JL4). For engaging
the right coronary artery, a size larger, the JR5 tends to engage
the right coronary artery better than the JR4.
The multipurpose A catheter can be used in either arm
for inferiorly directed takeoffs, while the multipurpose B cath-
eter can be more useful for horizontal and superior directed
takeoffs. A modified version of the multipurpose A catheter is
the Barbeau catheter. This catheter adds a 1358 primary curved
tip to the end of the typical multipurpose catheter, allowing
better engagement of the right coronary artery and bypass
grafts from the right arm approach.
Like the multipurpose catheter, others have been devel-
oped to engage both the right and left coronary ostium from
arm access without switching catheters. The Kimny curve diag-
nostic catheter (Schneider), invented by Ferdinand Kiemeneij,
has three curves: the primary curve is 1458, secondary curve is
908, and a tertiary curve is 1338 (Fig. 17.11). These curves allow
backup support from the opposing aortic wall while allowing
coaxial engagement of the left main, right coronary ostium, or
vein grafts. To engage a Kimny curve catheter into the left main
or right coronary artery, two general techniques can be used
(Figs. 17.12 and 17.13) (7). The first involves bringing the cath- Figure 17.12 Technique for catheterizing the left coronary artery in
eter tip down from the level of the ipsilateral coronary cusp to the PA view using the Kimny catheter. The catheter is advanced so
the ostium that is to be engaged. From there the catheter can that the tip enters the left sinus of Valsalva. (A) The catheter tip is
either be withdrawn and turned counterclockwise or clockwise rotated counterclockwise and withdrawn until it enters the left cor-
to engage the left coronary or right coronary, respectively. The onary ostium. (B) The body of the catheter is rotated counterclock-
second technique involves the same torque but instead of wise and advanced gradually until the tip rises to enter the left
coronary ostium. Source: From Ref. 7.
withdrawing the catheter, it is advanced forward against the
respective cusps to point the tip up and into the coronary
ostium, similar to the Amplatz catheter. Similarly shaped cath-
eters from other companies (Jacky, Tiger II, and Sarah catheters approach generally takes the same curve around the aortic arch
from Terumo; radialbrachial catheter from Cordis) also will as the femoral approach, typical catheters for bypass grafts can
engage both coronary ostium (8). be tried first (multipurpose for right bypass, JR4 or LCB for left
To engage aortocoronary saphenous vein grafts from a bypasses). If the JR4 or LCB is not working for the left-sided
transradial approach is more difficult than the typical trans- grafts, an Amplatz left, Judkins left, or multipurpose can be
femoral approach described previously (9). As the left radial tried as a second option. For the right radial approach, for
[15/7/010/22:31:46] [194–207]
Figure 17.13 Technique for catheterizing the right coronary artery

in the LAO view using the Kimny catheter. The catheter is advanced
so that the tip enters the right sinus of Valsalva. (A) The catheter tip
is rotated clockwise and withdrawn slightly until the tip enters the
right coronary ostium. (B) The body of the catheter is rotated
clockwise and advanced gradually until the tip rises to enter the
right coronary ostium. Source: From Ref. 7.
grafts originating on the right side of the aorta [usually that

have distal anastomoses in the right coronary artery (RCA)], Figure 17.14 Yumiko diagnostic catheter design allows selective
again, a typical approach with a multipurpose or IMA catheter cannulation of the left internal mammary artery via the right radial
can be used. For left-sided aortic anastomoses of saphenous approach. Source: From Ref. 11.
vein grafts, a JR4 or LCB will not have enough reach from the
right radial approach. In this circumstance, an Amplatz left
catheter can be very helpful.
To engage a LIMA from the right arm is difficult. The
Yumiko catheter (Goodman, Nagoya, Japan) was invented planes allowing for better filling of the ventricle with contrast.
specially for this purpose (Fig. 17.14) (10). It must be advanced The catheter has a straight shaft with a circular tip usually 1 cm
against the aortic valve so its tip points upward. It is then in diameter. The circular tip is a safety feature to minimize the
torqued into the left subclavian artery, at which point the chance of ventricular trauma and myocardial staining with
catheter is pulled back to take out slack in the aortic looped contrast as compared with a straight endhole catheter like the
portion of the catheter, allowing it to dive forward and engage multipurpose. Since the contrast is focused over multiple ejec-
the LIMA (Fig. 17.15). This catheter is not readily available in tion points, the catheter will usually sit relatively quietly in the
the United States, but the special LIMA1 catheter (Cordis) has a left ventricle, as compared with an endhole catheter, which will
similar shape and is available in the United States (Fig. 17.16). shoot backward with a power injection. Common variations on
Alternatively, in the absence of a specially designed catheter, an the typical pigtail design include 1458 and 1558 angled shafts to
angled glide wire (Terumo) can be manipulated from an IM allow the catheter to favor a midventricle position rather than
catheter from right arm access to aim up the left subclavian favoring the inferior wall, which can be common with a straight
(12). The wire is extended out into the left subclavian and the shafted pigtail (Fig. 17.17). A midventricle position of the
IM catheter is used to track over it, past the takeoff of the LIMA. catheter is important to allow for even angiography and to
After that, the wire is removed and the catheter is flushed and avoid ventricular ectopy. Pigtails are also available with radio-
hooked to the manifold. Test injections are given as the IM paque markers along the shaft to assist in quantitative angiog-
catheter is pulled back. If the IM catheter cannot initially aim raphy. As mentioned above, the “B” type multipurpose
the J-wire into the left subclavian, a Judkins left catheter can be catheter can be used for ventriculography as well. However,
used to send an exchange length guide wire into the left since the catheter has fewer sideholes and a straight tip, there is
subclavian, at which point the Judkins left catheter can be more chance of myocardial staining, ventricular ectopy, and
exchanged for an IM for selective angiography. A similar catheter movement as compared with a pigtail. In general,
method can be used to engage the RIMA from left arm access. lower flow rate is recommended through a multipurpose
catheter to minimize these complications, but this can lead to
poor ventricular visualization. For these reasons, we recom-
VENTRICULAR ANGIOGRAPHIC CATHETERS mend a standard angled pigtail as the first choice for left
Left ventriculography remains a standard component of a ventriculography.
diagnostic catheterization. A pigtail catheter has an endhole Right ventriculography is less commonly done in adult
plus multiple sideholes that will inject contrast in multiple catheterization but can be useful in the adult patient with
[15/7/010/22:31:46] [194–207]
Figure 17.15 (A) The technique of advancing the YUMIKO-LITA catheter into the LITA via the ascending aorta. The catheter is advanced to
the ascending aorta using a guide wire (left image). After the guide wire is removed, the catheter is rotated to direct the tip toward the left
subclavian artery. The tip easily selects the left subclavian artery by a slight clockwise rotation (second image). Pulling the catheter with a
slight counterclockwise rotation, the tip is advanced to the distal subclavian artery (third image). Further pulling of the catheter enables the tip
to be engaged into the LITA (right image). (B) The technique of advancing the YUMIKO-LITA catheter into the LITA via the descending aorta.
The catheter is advanced to the descending aorta using a guide wire. After guide wire removal, the catheter is turned, hanging the tip on the
side branch of the descending aorta at the celiac trunk or renal artery (left image). By pulling the catheter tip to be directed to the left (second
image), the tip is advanced toward the left subclavian artery (third image). Further pulling enables the tip to be engaged into the LITA (right
image). Source: From Ref. 11.
Figure 17.17 Straight and angled pigtail catheters. Abbreviation:

PIG, pigtail.
Figure 17.16 The LIMA1 catheter is used to engage the LIMA from
right radial access, similar to a Yumiko catheter.
atrium or ventricle with catheter manipulation. There is no

endhole so catheter recoil with injection is not an issue. It is
congenital heart disease, such as pulmonic stenosis, or if pul- inserted through the femoral vein and up the IVC with the use
monary artery angiography is warranted. The modern catheter of balloon flotation in an exact manner as a pulmonary artery
of choice for right ventriculography is a Berman angiographic flotation catheter can be inserted. If the right ventricular cham-
catheter developed by Dr Michael Berman. This is a sidehole ber is to be visualized the balloon is kept inflated to stabilize the
only catheter with a balloon flotation tip (Fig. 17.18). The catheter and a power injection is given. The Berman catheter
balloon flotation mechanism allows safe and easy passage can accept flow rates and PSI similar to the typical left
into the right ventricle without the risk of trauma to the right ventricular pigtail catheter. It can also be easily floated into
[15/7/010/22:31:46] [194–207]
PERIPHERAL DIAGNOSTIC CATHETERS

The design of peripheral diagnostic catheters is similar to that
of coronary catheters, and in some cases, coronary diagnostic
catheters are used for a wide variety of peripheral angiography.
There are also specialty-designed catheters for accessing chal-
lenging anatomy such as the carotid arteries and the visceral
abdominal vessels. Frequently, a nonselective angiogram is
used to visualize anatomic takeoffs of branch vessels, and
then more specialized catheters are used to engage for selective
diagnostic catheterization. Common peripheral angiographic
catheters are described below on the basis of the anatomic
location where they are used.
CAROTID
Typically, an angiogram of the aortic arch using a straight-
bodied pigtail catheter is used for initial information about the
location of the takeoffs of the carotid arteries and to see if there
Figure 17.18 Berman angiographic balloon floatation catheter. is any ostial stenosis that might make selective engagement of
diagnostic catheters dangerous. In most aortic arches, a JR4 can
be used to selectively engage and perform angiography of the
common carotids. As described in the above section for graft
anatomy, the JR4 can be advanced to the top of the ascending
aorta in the LAO 308 view, which lays out the aorta and its
branch vessels are not foreshortened. The JR4 can then be
advanced safely over the stiff glide wire to the proximal carotid
for selective angiography. Any catheter with one simple pri-
mary curve can be replaced for the JR4 for carotid angiography,
with the choice currently being very operator dependent.
Examples include an angled glide catheter (Terumo), IMA,
vertebral, or a headhunter catheter. A headhunter has a slight
secondary curve as well, but it still functions as a “simple-
curve” catheter to be used for normal types 1 and 2 arches and
its manipulation remains the same as the JR4. In cases of
tortuous anatomy or a type 3 aortic arch, preformed complex
curved catheters such as the Simmons or Vitek will be neces-
sary for selective engagement (13).
The Simmons sidewinder (SS) catheter, named after
Charles Simmons, typically comes in three sizes based on the
length of the catheter from the “knee” portion (secondary
curve) to the tip (Fig. 17.20) (14). The Simmons 2 is the most
common size used for carotid angiography and has a 4.5 cm
distance from knee to tip. The head of a Simmons catheter is
shaped like a shepherd’s crook, and before it is used, it must be
reformed and positioned in the ascending aorta. The preferred
technique (15) for reformation (Fig. 17.21) involves pushing the
catheter (with removed guide wire) over the aortic arch so that
its tip is in the ascending aorta looking caudally, and the
secondary curve of the catheter lies over the highest point of
Figure 17.19 Grollman pigtail catheter. the aortic arch. At this point the catheter is rotated clockwise
rapidly, which causes the catheter to buckle and loop over on
itself at the secondary curve resembling the shape of a scissor.
With continuing rotation the catheter tip will loop back over
the main pulmonary artery for selective pulmonary artery from the ascending to the descending aorta. At this point the
angiography. A second choice for right ventricle and pulmo- catheter’s secondary curve is pushed over the arch and is
nary angiography is the Grollman catheter, developed by allowed to rest on the aortic valve. With counterclockwise
Dr Julius Grollman. It is a modified pigtail catheter with a rotation the catheter is unwound and resumes its preformed,
reversed 908 secondary curve 3 cm from the catheter tip packaged shape. Alternatively, it can be reformed in the
(Fig. 17.19). This reverse curve allows it to engage the right ascending aorta against the aortic valve to its preformed
ventricle and to be advanced to the pulmonary artery if needed. shape like an Amplatz catheter (usually by following a stiff
This catheter is the second choice because it requires more wire that has prolapsed over the aortic valve). Once the
manipulation to advance into the right ventricle or pulmonary Simmons is reformed in the aortic root, the catheter should
artery. be rotated if needed so that the tip is pointing to the right of the
[15/7/010/22:31:46] [194–207]
Figure 17.20 General configuration and approximate dimensions of the distal portions of the Simmons “sidewinder” catheters I, II, and III.
Figure 17.21 Technical sequence for reforming the curve of the Simmons sidewinder catheter. (A) The potential knee of the catheter must
be at the most cephalad portion of the aortic arch. Clockwise rotation of the catheter is now initiated. (B) With rapid clockwise rotation, the
catheter will start to scissor at the knee. (C) With continued rotation, the scissoring is complete, as the tip flips over the arch to a position
adjacent to the distal catheter shaft in the proximal descending thoracic aorta. The reformation process is now complete. (D) The catheter is
advanced into the ascending aorta and rotated to open up the scissored configuration. Source: From Ref. 15.
patient and anterior in the aortic root. The catheter is then 1808 to look caudally so the catheter can be pulled back past the
pulled back until the tip engages the brachiocephalic trunk. takeoff of the brachiocephalic trunk. From here the tip is
This might require some clockwise or counterclockwise manip- rotated back 1808 to look cranial, and the catheter is pulled
ulation to successfully engage. Once the tip has engaged, it is backward until the tip pops into the left carotid artery. The
pulled back further, which will extend the tip cranially into the Simmons catheter can tend to loop on itself or get knotted, and
great vessel. This should always be done over a guide wire. A particular attention must be paid to avoid this complication.
stiff angled glide wire is ideal. The angled tip allows for The Vitek (Cook) catheter is a modified Simmons with a
directional advancement and the stiff body allows straighten- 908 reverse curve of the shepherd’s crook tip (Fig. 17.22). This
ing and advancement of the complex curved catheter. A softer allows for the catheter to more easily reform in the descending
guide wire would let the complex curve keep its shape and aorta or distal aortic arch, unlike the Simmons. It is placed over
would cause prolapse into the aorta with further advancement. a guide wire into the top of the descending thoracic aorta. As
For engagement of the left common carotid, the Simmons the guide wire is removed, it returns to its preformed shape. It
catheter is advanced without a guide wire to prolapse its is pushed proximally across the arch until it engages one of the
body back into the ascending aorta. From here the tip is rotated great vessels, where the tip will jump cranially into the vessel.
[15/7/010/22:31:46] [194–207]
Figure 17.22 The Vitek catheter has a similar hooked tip as the
Simmons catheter, but it has a 908 reversed curve, which makes it a
“push” catheter for engagement.
At this point, the catheter is pulled back straightening out the

tip and sending it further cranially. In the case of left carotid
angiography, no further manipulation is needed. If right
carotid angiography is needed, the Vitek catheter has selec-
tively engaged the brachiocephalic trunk. From this point a stiff
angled glide wire can be advanced into the appropriate vessel,
and the Vitek catheter will track over this for selective angiog-
raphy. The big difference between the Simmons and Vitek
catheter is location of reformation of the preformed shape. On
the basis of location, the Vitek becomes a “push” catheter to
engage the arch vessels while the Simmons becomes a “pull”
catheter to engage the vessels. Also, once the Vitek has engaged
the brachiocephalic trunk, the tip of the catheter is pointed Figure 17.23 Vertebral catheter.
cranially and slightly rightward, which allows easier wire
manipulation into the right carotid whereas the Simmons tip
will point slightly leftward, requiring slightly more manipula-
tion to engage the right common carotid. For these reasons, the
Vitek usually requires less manipulation than a Simmons cath- gentlest primary curve of 1508 (i.e., the tip is only angled 308
eter and is generally preferred as the first choice for difficult from a completely straight position) (Fig. 17.23).
arch anatomy.
For a bovine arch with the left carotid coming off low in
the brachiocephalic trunk, the Vitek catheter is best used to
SUBCLAVIAN AND UPPER EXTREMITY
selectively engage the left carotid. As described above, upon
Similar to carotid angiography, an aortic angiogram in the LAO
engagement of the brachiocephalic trunk, the Vitek tip natu-
308 to 458 view will give a nonselective view of the takeoff of the
rally points cranially and to the right. Once the Vitek is engaged
right and left subclavian arteries. As described in the section
in the brachiocephalic trunk, if the tip is already above the left
“Carotid,” a JR4 or other simple curved catheter can usually be
carotid takeoff, then the catheter is advanced forward until the
manipulated in most cases for selective left and right subclavian
tip moves down the brachiocephalic trunk. A stiff angled glide
angiography. In the case of tortuous or type 3 arches, a Vitek or
wire can then be used to selectively engage the left carotid. The
Simmons catheter is preferred to engage the left or right sub-
catheter can be pulled backward to advance the tip selectively
clavian. Once selective subclavian angiography is performed,
over the wire into the “bovine” left carotid. A Simmons catheter
runoff to the upper extremity can be visualized with panning of
requires significantly more manipulation into a bovine left
the angiographic table. For more selective angiography of the
carotid as the initial tip looks rightward and should not be
upper extremities, a simple curved catheter like a JR4 or IMA
used.
can be used, and advanced over a wire more distally in the arm.
If a complex curve catheter is used for initial angiography, an
exchange length Wholey or glide wire can be placed distally in
the arm. Following this, the original diagnostic catheter is
VERTEBRAL
walked out, and can be exchanged for a straight or angled
Angiography of the vertebral arteries is considered part of a
glide catheter or a multipurpose catheter.
complete cerebral circulation study (along with carotid angiog-
raphy with anterior cerebral runoff) to fully evaluate the pos-
terior cerebral circulation. Again, the first choice for selective
vertebral angiography remains a simple curve catheter with the RENAL
JR4 being the first choice. Alternatives include the IMA cathe- Renal angiography begins with a nonselective descending aor-
ter, angled glide catheter, or a vertebral catheter, based on tic angiogram (utilizing a straight pigtail catheter) at the level of
operator training and preference. The vertebral catheter has the the takeoff of the renal arteries. The renal arteries typically
[15/7/010/22:31:46] [194–207]
Figure 17.25 Tip of Sos Omni catheter. It has a similar hooked

shape as a Simmons catheter, but it has a much tighter radius of
curvature, allowing for easier reformation of its primary shape.
Figure 17.24 Cobra catheter.
aortic wall. If a renal artery has an inferiorly directed takeoff, a

branch off of the descending aorta at the interspace between the Cobra catheter might not be able to successfully engage. In this
L1 and L2 vertebrae. The sideholes of the pigtail catheter thus case, an IM catheter, which has a more acutely angled tip,
will be at the level of the renal ostia. The right renal artery might be able to engage. If the takeoff has an extremely inferior
usually comes off at a slightly higher level than the left renal takeoff, a shepherd’s crook-shaped catheter (“hooked” end
artery. The PA angiogram will give useful information regard- similar to a Simmons catheter as described but with a smaller
ing the orientation of the renal ostia or anatomic variants such overall curve radius), such as the Sos Omni (Angiodynamics),
as an accessory renal artery. Since many patients whose renal can be useful (Fig. 17.25). The Sos is named after its inventor,
arteries are being assessed have renal insufficiency, the pigtail Tom Sos. Similar to a Simmons catheter, this hooked catheter
angiogram can be done with minimal contrast (only 10 cc requires a pull technique to engage the renal ostia. The catheter
needed at high flows), and the operator can proceed with is brought above the level of the renal ostium over a wire. As
selective angiography. Typically, the renal arteries have a fairly the wire is removed, the catheter will return to its prepackaged
orthogonal takeoff of the lateral aspect of the aorta. The right shape in the abdominal aorta since it has such a tight curve. The
renal artery has a slightly anterior takeoff as well. The JR4 catheter is then manipulated either clockwise or counterclock-
catheter again is very useful as the first choice to engage the wise to face the lateral wall of the aorta. The catheter is then
laterally directed renal ostia. Alternatively, a Cobra catheter brought down and when it is thought to hook a renal ostium, a
(invented by Melvin Judkins) can be used to selectively engage test injection is given. Another method, called the “flick” tech-
the renal arteries. The shape of this catheter looks like a Cobra nique, allows engagement of a renal artery without multiple
head or question mark and is similar in shape to the LCB injections of contrast (Fig. 17.26). This method involves reform-
catheter described previously. The Cobra has three curves. The ing the Sos Omni in the abdominal aorta below the level of the
primary curve has a 908 tip angled from the shaft of the renal arteries. A soft tipped straight guide wire like the Bentson
secondary portion of the catheter. Along with about a 708 guide wire is left in the preformed catheter with 1 cm extending
secondary curve in the same direction, the tip points slightly from the catheter tip. Because the tip of the Bentson guide wire
downward to be able to hook into visceral ostia (Fig. 17.24). is very floppy, the catheter remains in its preformed shape
Finally, a very broad tertiary curve is angled in the opposite despite having the guide wire extending from its tip. This
direction of the primary and secondary curves. This reversed floppy tip also allows for minimal aortic trauma. The catheter
tertiary curve allows contralateral support from the opposite will engage the renal ostium with a push technique. The tip of
[15/7/010/22:31:46] [194–207]
Figure 17.26 “Flick” technique for engaging the renal artery without scraping the catheter tip against the wall of the aorta. (A) Sos catheter
advanced below the level of the renal artery with a soft tipped guide wire. (B) Guidwire is withdrawn into the catheter, leaving 1 cm extended.
(C) The Sos catheter is advanced. (D) As the catheter is advanced, the guide wire will flick laterally into the renal artery, at which point the
catheter can be advanced. Source: From Ref. 16.
the guide wire remains parallel to the aortic wall until the reform in the descending aorta. The catheter is pulled back and
catheter is pushed high enough that the wire will flick laterally rotated with its tip toward the anterior aorta (tip looking
into the selective renal artery. The catheter can then be laterally in the lateral fluoroscopic angulation) until it is seen
advanced to the vessel ostium and the wire can be pulled out. and felt to hook into a visceral ostium.
VISCERAL LOWER EXTREMITY

Angiography of the major abdominal aortic branches (celiac Lower extremity angiography usually begins with a nonselec-
trunk, superior mesenteric artery, and inferior mesenteric tive angiogram with a pigtail catheter in the distal descending
artery) begins with a nonselective abdominal aortic angiogram aorta above the level of the iliac bifurcation. Nonselective
with a nonselective flush catheter, like a pigtail. Angiography is angiography is performed, and the table is panned to view
done in the lateral orientation to identify the origin of the runoff from the iliac artery all the way to the below the knee
appropriate vessel, as they have takeoffs from the anterior runoff. If a stenosis is visualized, more selective angiography
aorta. The tip of the pigtail should be placed above the level can be performed with catheters selectively placed in any of the
of T11 so the celiac trunk can be visualized at the level of T12. vessels (contralateral to the access sheath) or through the sheath
Runoff should visualize the takeoff of the SMA at the level of in the femoral artery (to image the ipsilateral vessels). The
L1 and the IMA at the level of T3. After nonselective angiog- common iliac arteries are usually spaced about 908 apart, so a
raphy is performed, selective angiography is usually performed catheter with around a 908 tip, like a JR4 or IMA, is the primary
with a simple curved catheter like a JR4 or IMA. If the JR4 or choice to selectively cannulate contralateral vessels. If the angle
IMA is unable to hook the vessel, a more complex curved between the iliac arteries is <908, a hook-shaped catheter head
catheter like an Sos or Cobra can be helpful. To hook the can be more useful to hook the ostium of the ipsilateral iliac.
appropriate vessel, the catheter is positioned at a level above Examples of this include the Sos, shepherd’s hook (SHK), or
the artery that is intended to be studied with a guide wire. The RIM catheters (Fig. 17.27). Once a wire is advanced from the
guide wire is removed allowing the complex curved catheter to diagnostic catheter up and over, the diagnostic catheter is
[15/7/010/22:31:46] [194–207]
4. Isshiki T, Yamaguchi T, Tamura T, et al. Percutaneous angioplasty

of stenosed gastroepiploic artery grafts. J Am Coll Cardiol 1993;
22:727–732.
5. Tanimoto Y, Matsuda Y, Fujii B, et al. Angiography of right
gastroepiploic artery for coronary artery bypass graft. Cathet
6. Schoonmaker FW, King SB III. Coronary arteriography by the
single catheter percutaneous femoral technique. Experience in
6,800 cases. Circulation 1974; 50:735–740.
7. Shibata Y, Doi O, Goto T, et al. New guiding catheter for transrad
PTCA. Cathet Cardiovasc Diagn 1998; 43:344–351.
8. Kim SM, Kim DK, Kim DI, et al. Novel diagnostic catheter
specifically designed for both coronary arteries via the right
transradial approach. A prospective, randomized trial of Tiger II
vs. Judkins catheters. Int J Cardiovasc Imaging 2006; 22:295–303.
9. Burzotta F, Trani C, Hamon M, et al. Transradial approach for
coronary angiography and interventions in patients with coronary
Figure 17.27 The SHK and RIM catheter can be used to access a bypass grafts: tips and tricks. Catheter Cardiovasc Interv 2008;
contralateral iliac artery in the “up and over” approach. Abbreviation: 72:263–272.
SHK, shepherd’s hook. 10. Kim MH, Cha KS, Kim HJ, et al. Bilateral selective internal
mammary artery angiography via right radial approach: clinical
experience with newly designed Yumiko catheter. Catheter
Cardiovasc Interv 2001; 54:19–24.
11. Miyake Y, Inoue T, Morooka S, et al. A novel method for
walked out and a long flexible sheath is placed up and over to angiography of the left internal thoracic artery from a right arm
the contralateral iliac artery. At this point, more selective approach using a YUMIKO-LITA catheter. Am J Cardiol 2002;
89:984–986.
angiography can be performed through the sheath itself or
12. Zheng H, Pentousis D, Corcos T, et al. Bilateral internal mammary
through a straight glide catheter inserted distally from the
angiography through a right radial approach: a case report. Cathet
guide wire to the area of interest. Cardiovasc Diagn 1998; 45:188–190.
13. Simmons CR, Tsao EC, Thompson JR. Angiographic approach to
the difficult aortic arch: a new technique for transfemoral cerebral
REFERENCES angiography in the aged. Am J Roentgenol Radium Ther Nucl
1. Devlin G, Lazzam L, Schwartz L. Mortality related to diagnostic Med 1973; 119:605–612.
cardiac catheterization. The importance of left main coronary 14. Grable GS, Smith DC. The use of the Simmons “Sidewinder”
disease and catheter induced trauma. Int J Card Imaging 1997; catheter in percutaneous transluminal angioplasty of the renal
13:379–384; discussion 85–86. arteries. Radiology 1980; 137:541–543.
2. Tilkian AG, Daily EK. Cardiovascular Procedures: Diagnostic 15. Smith DC, Simmons CR. The quick aortic turn: a rapid method for
Techniques and Therapeutic Procedures. St. Louis: Mosby, 1986. reformation of the Simmons sidewinder catheter. Radiology 1985;
3. Myler RK, Boucher RA, Cumberland DC, et al. Guiding catheter 155:247–248.
selection for right coronary artery angioplasty. Cathet Cardiovasc 16. Sos TA, Trost DW. Tips and Tricks of RAS for High-Risk Patients.
Diagn 1990; 19:58–67. Endovascular Today 2003.
[2/7/010/8:43:31] [208–228]
18
Coronary imaging: angiography, CTA, MRA

Joel A. Garcia and John D. Carroll
INTRODUCTION lesions, thrombus burden, and vessel spasm can be evaluated

The central diagnostic goal of contemporary coronary arterial with SCA (2,3). Access to the aorta is commonly obtained by
imaging is to completely visualize the coronary artery tree, to cannulation of the femoral artery although the radial, brachial
understand the anatomy, and to recognize and characterize and axillary approaches have been also used. Radial arterial
pathologic findings. Beyond the diagnostic role, imaging is access is particularly common outside the United States. Vas-
used to plan and execute treatment using the recognition and cular access is described in detail in chapter 8. Shape specific
quantification of a variety of vessel and lesion features prior to catheters are then advance through the aorta to the ostium of the
interventions. The therapeutic planning goals are to develop an coronary arteries over a guidewire. Once the wire is removed
equipment strategy for each lesion to be treated. Likewise an and the catheter is engaged the injection of contrast with the
imaging strategy is developed before interventions that allows subsequent radiographic acquisition results in an angiographic
for a working view to provide real-time visualization during the image. Hemodynamic parameters are constantly recorded
procedure. Finally, at the completion of the intervention, coro- before, during, and after injections. Diagnostic catheters are
nary imaging is used to confirm and quantify the therapeutic the focus of chapter 17.
changes to the coronary arteries and detect any complications. The two-dimensional (2-D) nature of SCA requires that
While the imaging technologies of computed tomo- multiple single-view angiographic acquisitions are performed
graphic angiography (CTA) and magnetic resonance angiogra- in various projections. These views are obtained by rotating the
phy (MRA) are now available and used clinically, catheter- imaging system around the patient who lies supine on a
based standard coronary angiography (SCA) remains the gold radiolucent table (2,3). The imaging system can be rotated
standard. SCA is not only widely available but also provides from left anterior oblique (LAO) to right anterior oblique
the highest spatial and temporal resolution for coronary imag- (RAO) with an angulation toward the head (cranial) or an
ing in 2009. SCA provides reliable angiographic information angulation toward the legs (caudal). While a wide array of
with a vast clinical experience over multiple decades of use. It potential combinations of rotation and angulation is possible,
is routinely used in the clinical setting to determine the appro- several standard approaches result in a comprehensive and
priateness of percutaneous coronary interventions (PCI), coro- mostly complete visualization of the coronary anatomy.
nary artery bypass grafting (CABG), or medical therapy. The methods of performing coronary angiography have
Coronary angiography, first performed by Dr. Mason Sones evolved substantially. Smaller catheters are currently used
in 1959, has been a cornerstone in the development of the resulting in minimization of vascular risk, early ambulation
concept of transluminal angioplasty and revascularization and discharge (3). Images were traditionally stored on 35-mm
therapies. This includes the work of Charles Dotter who cinefilm but now are almost exclusively digital recordings
performed the first transluminal angioplasty in 1964, René stored on local hard drives, servers, and occasionally compact
Favaloro who performed the first CABG in 1967, and Andreas disk (CD) or digital video disk (DVD) media.
Gröentzig who performed the first percutaneous transluminal
angioplasty in 1977 (1–3). The use of diagnostic coronary Coronary Anatomic Considerations
angiography has significantly increased in the last decade The individual variability of the coronary anatomy requires
both with the high prevalence of coronary artery disease that the acquisition technique and technology allow various
(CAD) in the industrialized world as well as an increase in combinations of angulation and rotation that result in a gantry
prevalence in the remainder of the world. This chapter will position that acquires a clinically useful image. Multiple gantry
focus on the traditional catheter-based coronary angiography positions are used in an effort to minimize the imaging inac-
and two revolutionizing noninvasive technologies that also curacies resulting from viewing a three-dimensional (3-D)
define the coronary anatomy, which are CTA and MRA. structure in two dimensions. These imaging inaccuracies and
misrepresentations are mostly related to vessel overlap, ostial
lesion characterization, and foreshortening (Fig. 18.1). Overlap
Standard Coronary Angiography results from the superimposed image of one vessel on another.
Coronary angiography clearly delineates the course and size of The same concept applies for an ostial lesion that may be
the coronary arteries, identifies anomalies, and provides infor- visually obscured by the main vessel. Foreshortening results
mation on the location, characteristics and degree of obstruc- from the relationship of a vessel segment’s longitudinal path in
tions. It provides information on coronary origin, vessel size, relation to the path of the X-ray beam. This can result in a
artery pathway, branches, lesion presence, collateral circulation, projection image that may minimize the true length of a vessel
and myocardial bridging. Aneurysmal segments, thrombotic segment as well as a coronary lesion.
[2/7/010/8:43:31] [208–228]
CORONARY IMAGING: ANGIOGRAPHY, CTA, MRA 209
Figure 18.1 Standard angiography imaging inaccuracies and 3-D modeling. (A) Representation of overlap as a common imaging limitation;
(B) ostial lesion representation; (C) 3-D modeling concept; and (D) vessel foreshortening representation. Source: From Ref. 73.
Standard Coronary Angiography Limitations technologies transform the 2-D images into 3-D models and
Although SCA has significantly improved over the past 20 years reconstructions. These will be discussed subsequently.
in terms of the quality of the image, postprocessing enhance-
ment, display on high-quality monitors, and tools for replay
and lesion quantification, it is still plagued with limitations IMAGE INTENSIFIERS AND FLAT-PANEL
inherent to 2-D projection images of the opacified inner diam- DETECTORS
eter, that is, lumen, coronary arteries, and the obvious risk of Imaging-Related Coronary Fundamentals
exposure to contrast media and radiation (Table 18.1) (4,5). and Limitations
These limitations and the so called “luminology” evaluation The development of techniques for imaging the vasculature
of the vessel have resulted in the need to have alternative over the last 100 years has revolutionized the understanding
technologies that allow for a more precise characterization of and treatment of patients with cardiovascular, cerebrovascular,
vessel anatomy, lesions, and stents. Intravascular ultrasound and peripheral vascular disease (PVD). The history of vascular
(IVUS) is one widely available technology that complements imaging began shortly after the development of radiography
SCA in the need for a more precise coronary arterial evaluation. when, in 1896, Hascheck and Lindenthal injected and imaged
Another group of technologies enhances the projection images the vasculature of an amputated hand with a mixture of salts.
to highlight certain structures like coronary stents and other Soon thereafter, femoral and cerebral arteriography using
sodium bromide and strontium bromide injections were
described in 1924 and 1927. With the development of less
Table 18.1 Standard Coronary Angiography Imaging and Safety toxic contrast agents and better imaging equipment, these
Limitations initial experiences resulted in surgical interventions to treat
peripheral and cerebrovascular disease and formed the foun-
Standard angiography Standard angiography dation for the field of interventional radiology (1).
Coronary imaging limitations Patient safety limitations
Contemporary medicine has evolved along with current
Vessel foreshortening Invasive procedure imaging techniques and radiographic equipment. Image inten-
Vessel overlap Contrast exposure sifiers (II) have evolved into flat-panel detectors that allow for
Ostial and bifurcation lesion evaluation Radiation exposure the most advanced vessel evaluations. Images are digitally
Lesion eccentricity evaluation
enhanced and the equipment software constantly adjusts
[2/7/010/8:43:31] [208–228]
settings to provide a superior evaluation. The traditional 2-D the study. Several conditions however do present a relative
planar imaging evaluation is now complemented by advanced contraindication to angiography. Active bleeding, severe ane-
imaging systems like biplane angiography, rotational angiog- mia (hemoglobin <8 gm/dL), unexplained fever, active or
raphy (RA), single- and dual-axis RA, computed tomography untreated infections, severe electrolyte imbalance, digitalis tox-
(CT) angiography, and magnetic resonance (MR). Most of icity, uncontrolled systemic hypertension, previous serious
these techniques and technologies allow for representation of contrast reaction, severe agitation and acute renal failure are
vessels in 3-D utilizing workstations with advanced graphics relative contraindications. Others include ongoing stroke,
capabilities. decompensated congestive heart failure (CHF), intrinsic or
iatrogenic coagulopathy [international normalized ratio (INR)
> 2.0] and active endocarditis (3,11). Individuals at risk for
3-D Acquisition and 3-D Visualization developing serious complications from coronary angiography
The display and quantification of 3-D vessel properties has
are also important to recognize before the procedure.
been a recent development in coronary imaging. Modeling
and/or reconstructive techniques based on SCA have been a
significant step forward in visualizing vessels in 3-D and some Coronary Angiography in Vulnerable Populations
approaches have become commercially available (Fig. 18.1C) There is an increased risk of complications in general in those
(6–8). The ability to produce 3-D coronary models and recon- individuals who exceed 70 years of age, have congenital heart
structions has led to quantification of clinically relevant 3-D disease, obesity, cachexia, uncontrolled glucose intolerance,
vessel properties such as tortuosity, bifurcation angles, take-off arterial oxygen desaturation, severe chronic obstructive lung
angles, vessel ostium orientation, and accurate segment length disease, and renal insufficiency with creatinine greater than
determination. 1.5 mg/dL (3,11). There is an increase risk of cardiac compli-
The coronary tree rests on the epicardial surface of the cations among patients with three-vessel CAD, left main (LM)
heart, moves with the underlying myocardium, and as a result lesions, New York Heart Association (NYHA) class IV, signif-
changes in 3-D shape. Attempts to quantify the specific icant mitral or aortic valve disease, ejection fraction (EF) < 35%,
dynamic behavior of a coronary vessel throughout the cardiac high-risk exercise treadmill test (ETT), pulmonary hyperten-
cycle, that is, 4-D shape changes, using traditional SCA have sion, and pulmonary artery wedge pressure (PCWP) in excess
previously been unsuccessful. The clinical relevance of 4-D of 25 mmHg (3,11). Another group of patients is at particular
shape changes have come to light with problems such as risk of vascular complications including those with anticoagu-
stent-induced conformational changes in coronary artery lation or bleeding diathesis, uncontrolled systemic hyperten-
shape, that is, straightening, and the risk of stent fracture sion, severe PVD, recent stroke and severe aortic insufficiency
from repetitive deformation. A multimodality 3-D vessel quan- (3,11).
tification algorithm will likely be part of future endovascular
therapies.
Fundamentals
Accurate diagnosis of CAD requires multiple injections in
CORONARY CONTRAST DELIVERY METHODS various views to produce clinically useful images of all coro-
Optimizing coronary angiography requires reliable, adequate, nary segments by minimizing overlap and foreshortening (9).
and steady injections of contrast media during image acquisi- Traditionally, invasive coronary angiography has been per-
tion. Furthermore, the use of right coronary artery (RCA) for formed with radiographic equipment that provides a simple
volumetric reconstruction requires a longer coronary contrast 2-D projection image representation of a patient’s more com-
injection. The traditional manifold allows for the injection of plicated 3-D anatomy. This “flattening” of a 3-D image results
contrast media but suffers from the inherent limitations of in the generation of a final image that may be limited and
being unable to provide a standard amount of flow (mL/sec). inaccurate. To minimize these limitations, invasive cardiolo-
This is also coupled to the fact that the maximal contrast gists typically acquire 6 to 10 diagnostic views of the coronary
volume (mL) is fixed with the size of the syringe attached to arteries. In addition to the limitations of intravenous contrast,
the manifold. Newer power injectors allow for the combination ionizing radiation, and procedural time, each angiographic
of a variety of flows and volumes that can be predetermined by view is subjectively chosen to best display an individual
the operator (9,10). The safety of these longer direct coronary patient’s coronary arteries. As it has been previously demon-
injections (<7.1 seconds) has been evaluated (10). strated, the combination of 2-D imaging limitations and this
trial-and-error technique results in some unrecognized imaging
inaccuracies (Table 18.1) (12–15). Nonetheless, SCA remains the
STANDARD CORONARY ANGIOGRAPHY gold standard for the visualization of coronary artery anatomy
Indications and for the detailing of CAD.
The American College of Cardiology (ACC)/American Heart
Association (AHA) Task Force has delineated the indications
for coronary angiography in patients with known or suspected
Traditional Coronary Angiographic Views
Not all of the many potential views of a patient’s coronary tree
CAD (Table 18.2) (3,9,11). Please refer to chapter 5 for a full
are necessary for an adequate study; rather a reasonable com-
review of cardiac catheterization patient selection including
bination of views must be acquired that adequately display the
patients with stable angina.
patient’s coronary tree and abnormalities. The easy positioning
of the gantry allows for a wide variety of combinations of
Contraindications rotation and angulation. Rotation is defined as the position of
There are no absolute contraindications for coronary angiog- the flat panel or II toward the right side of the chest (RAO) or
raphy other than an individual who is not willing to consent to the left side of the chest (LAO). Angulation refers to the
[2/7/010/8:43:31] [208–228]
Table 18.2 CTA/MRCA Appropriateness Criteria

Coronary artery
disease pretest HRF ECG Cardiac Prior to Coronary
probability NIT ECG Exercise changes markers CTA Risk CTA MR angiography
Chest pain syndrome Interm - Interp Able - - - - U (5) I (2) -
Interm - Interp Unable - - - - A (7) I (2) Class I (if
refractory
to medical
therapy)
High - - - - - - - I (2) I (1) Class I
Coronary anomaly suspected/ - - - - - - - - A (7–9) A (8) -
congenital heart disease/ Low Y - - N - - - U (5) - Class I (if no
new-onset CHF (CTA only) HRF IIb)
Acute chest pain/unstable angina Interm - - - N Neg - - A (7) U (6) -
High - - - N Neg - - U (6) - Class I
High - - - Y Pos - - I (1) I (1) Class I
Interm - - - Y Pos - - U (4) - Class I
Triple rule out/NSTEMI - - - - - - - - U (4) - -
Asymptomatic - - - - - - - Low I (1) - Class III

- - - - - - - Mod I (2) - Class III
- - - - - - - High U (4) - Class III
Uninterpretable test (exercise - - - - - - - - A (8)
treadmill test, perfusion, stress
echo)
Moderate-severe ischemia - - - - - - - - I (2) -
Risk assessment (asymptomatic) - - - - - - <2 yr High I (2) - Class III
(NOD)
- - - - - - Ca score High I (3) - Class III (as
> 400 screening
test)
Preoperation evaluation (noncardiac
surgery) - - - - - - - Interm I (1) -
Low-risk surgery - - - - - - - Interm U (4) -
High-risk surgery
Postrevascularization - - - - - - - - - - -
CABG (CPþ) - - - - - - - - U (6) I (2) Class IIb
PCI (CPþ) abrupt closure? Stent - - - - - - - - U (5) I (1) Class I
thrombosis? - - - - - - - - I (2) - Class IIb
CABG (no CP) - - - - - - - - I (2–3)- - Class IIb
PCI (no CP) ISR - Y - - - - - - - - Class I
PCI < 9 mo - Y - - - - - - - - Class IIb
CORONARY IMAGING: ANGIOGRAPHY, CTA, MRA
CABG (CPþ)/constant angina - - - - - - - - - Class III

CABG (CP) no ischemia
(Continued )
211
212
[2/7/010/8:43:31] [208–228]
Table 18.2 CTA/MRCA Appropriateness Criteria (Continued )

Coronary artery
disease pretest HRF ECG Cardiac Prior to Coronary
probability NIT ECG Exercise changes markers CTA Risk CTA MR angiography
Nonspecific CP - Y - - - - - - - -
Recurrent hospitalization - - - - - - - - - - Class I
Equivocal findings - - - - - - - - - - Class IIb
NIT
After Q-wave myocardial infarction or
NQWMI - - - - - - - - - - Class I
Active ischemia - - - - - - - - - - Class I
Presurgical therapy (MR, VSD) - - - - - - - - - - Class I
Hemodynamic instability - - - - - - - - - - Class IIa
MI due to embolism, arthritis, trauma, - - - - - - - - - - Class IIa
metabolic disease, spasm - - - - - - - - - - Class IIb
MI survivors (ejection fraction < 40%, - - - - - - - - - - Class IIb

CHF, PCI, CABG, arrhythmia) - - - - - - - - - - Class IIb
CARDIOVASCULAR CATHETERIZATION AND INTERVENTION
Delayed PCI for infarct related artery

Left main or 3-vessel disease
Recurrent ventricular tachycardia or all
NQWMIs
Abbreviations: CTA, computed tomography angiography; MRCA, magnetic resonance coronary imaging; HRF, high-risk features on noninvasive testing; NIT, noninvasive testing; Interm,
intermediate; Interp, interpretable; Ca, calcium; Neg, negative; Pos, positive; NOD, nonobstructive disease; Y, yes; N, no; I, inappropriate; A, appropriate; U, uncertain; PCI,
percutaneous coronary intervention; CABG, coronary artery bypass graft; ISR, in-stent restenosis; NQWMI, non–Q wave myocardial infarction; CHF, congestive heart failure; PCI,
percutaneous coronary intervention; LM, left main; 3VD, three-vessel disease; MR, magnetic resonance; MI, myocardial infarction.
[2/7/010/8:43:31] [208–228]
Figure 18.2 Rotational angiographic images. (A) Catheter position in relation to the spine. The tip of the catheter to the left of the spine and
the catheter in the form of an inverted “U” shape suggest LAO rotation. The tip of the catheter to the right of the spine and the catheter shape
in an overlapped form suggest RAO rotation. The tip of the catheter on top of the spine suggests no significant RAO or LAO rotation. This
concept also applies to the RCA injection as shown on the dual injection performed in (B). (B) RA (cranial) of a dual injection the LCA and the
RCA. (C) RA of the LCA. Note the relationship between the angiographic image obtained for both vessels and the position of the gantry below
(D). From left to right the angiographic images and the gantry representations rotate from LAO to RAO with 258 of cranial angulation. The
presence of the diaphragm on all images suggests that they are cranial in angulation. Abbreviations: RA, rotational angiography; LCA, left
coronary artery; RCA, right coronary artery; LAO, left anterior oblique; RAO, right anterior oblique.
position of the flat panel or II toward the head of the patient of the angiogram. Any change in gantry angulation or rotation
(cranial) or toward the legs (caudal). The rotation (LAO vs. is reflected by a change in the image. Operators are required to
RAO) of a view may be recognized from the image itself by the be very proficient in real-time image interpretation and in the
position of tip the catheter in relation to the spine or aorta or creation of a mental imprint of the 3-D image while acquiring
based on the shape of the catheter (Figs. 18.2 and 18.3). In LAO only 2-D information. Over time the use of coronary angiog-
the catheter shape is that of an inverted “U” (Figs. 18.2 and raphy has resulted in the recognition of views that are usually
18.3A, B, C), while the shape of the catheter in RAO shows it adequate to display certain coronary segments. While this
overlapped on itself (Figs. 18.2 and 18.3E, F, G, H). The tip of “expert-recommended” approach has evolved from the prac-
the catheter to the right of the spine on the screen suggests RAO tice of angiography no scientifically based analysis of their
(Fig. 18.3E, F, G, H, I). The tip of the catheter to the left of recommended views has been performed. It is important to
the spine on the screen suggests LAO (Figs. 18.2 and 18.3A, B, recognize that the variability in anatomy from patient to patient
C, K, L). The presence of the diaphragm on the image is usually makes the recommended views only a guide and the angiog-
seen in cranial shots (Fig. 18.3H, I, J, K, L). rapher must individualize the views acquired to produce a
The technique used to acquire images impacts the accu- clinically adequate angiographic study. A detailed description
racy and reproducibility of traditional angiography. To mini- of the coronary anatomy is the focus of chapter 16. Detailed
mize imaging artifacts introduced by 2-D acquisitions, knowledge of the coronary anatomy is mandatory to perform
traditional angiography depends on at least two views which optimal coronary image acquisitions.
are orthogonal to the vessel segment of interest. This principle RAO caudal views are used to visualize the LM, proximal
led to the development and widespread use of biplane angio- left anterior descending artery (LAD), and proximal circumflex
graphic equipment. (Figs. 18.3E, F, G and 18.4F). RAO cranial views visualize the
Traditionally a series of screening views have been used mid- and distal LAD, without overlap of septal or diagonal
by operators. These screening views are ultimately comple- branches, and the distal posterolateral (PL) branches
mented by additional views based on real-time interpretation (Figs. 18.3H, I, J and 18.4B, M, N). The LAO cranial views
[2/7/010/8:43:31] [208–228]
Figure 18.3 Dual-axis rotational coronary angiography of the left coronary artery. Note the different views and the lesion characterization of
the mid-LAD obtained by a single injection. Several different views of the mid-LAD lesion are shown with an arrow. This set of views also
shows the relationship of the spine and the diaphragm based on angulation and rotation of the gantry. Note the eccentricity of the plaque
shown by the arrows. (A, B) LAO caudal, (C) posteroanterior caudal, (D, E) RAO caudal, (H, I) RAO cranial, and (K, L) LAO cranial.
Abbreviations: LAD, left anterior descending artery; LAO, left anterior oblique; RAO, right anterior oblique.
visualize the mid- and distal LAD in an orthogonal projection foreshortening and overlap with surrounding vessels. This
(Fig. 18.3K, L). LAO caudal views visualize the LM, ramus scientifically based evaluation validated current coronary
intermedius and the proximal circumflex (Fig. 18.3A, B, C). If angiographic knowledge and clinical practice but also allows
uncertainty remains the supplemental views include poster- extension into challenges of optimizing bifurcation regions
oanterior (PA), lateral 608 to 908 with cranial or caudal angu- (Figs. 18.5 and 18.6).
lation. The RCA views include LAO to visualize the proximal
portion (Fig. 18.4A, D, G, J), RAO cranial for the posterior
descending artery (PDA) and PL branches, RAO 308 and lateral
Limitations of Standard Coronary Angiography
The errors in image acquisition, evaluation and interpretation
for the mid segment (3,9,16,17).
can have a significant impact on the management strategy of
CAD (3,19). Ideally, a systematic approach aims at minimiz-
Universal Optimal View Map ing these inaccuracies and potentially avoids impacting clin-
Recognizing views that target specific vessel segments has ical outcomes. Technical issues are generally mastered with
evolved with the practice of angiography. An approach to experience. Inadequate vessel opacification due to conditions
scientifically validate optimizing views has been studied like aortic insufficiency, streaming, diluted contrast, compet-
using a in-depth analysis of a large patient population (18). A itive flow, anemia, noncoaxial catheters can result in image
database of vessel reconstructions segment by segment results interpretation limitations. Coronary angiography outlines the
in the creation of an optimal view map, that is, a graphic lumen of the vessel but is unable to provide wall thickness
representation of gantry position and the quantification of a information, visualization of diffuse disease, quantification of
key visualization measurement to be optimized, such as vessel nominal vessel diameter in the presence of plaque, and the
[2/7/010/8:43:31] [208–228]
Figure 18.4 Set of images showing various common angiographic findings. (A) RCA vessel tortuosity (black arrow); (B) left main coronary
dissection and zoomed image (arrow shows oblique line to the right of the catheter); (C) eccentric mid-LAD lesion (arrow) and zoomed image;
(D) RCA proximal ectasia (arrow); (E) overlap of the proximal LAD (black arrow), angulation and foreshortening of the mid-Cx (white arrow);
(F) concentric distal Cx lesion (black arrow) and RCA collaterals (white arrow); (G) complete occlusion of the mid-RCA (black arrow);
(H) severe ostial right ventricular marginal lesion (black arrow) and moderate posterior descending artery ostial lesion (white arrow). There is
also a mild RCA atrioventricular continuation lesion which then constitutes a bifurcation lesion (in circle); (I) ostial RCA catheter induced
spasm (black arrow); (J) RCA contrast streaming simulating dissection that normalized with appropriate injection flow and volume (black
arrow); (K) mid-RCA thrombus (black arrow points to contrast hang out). Please note that the white arrow points toward mild calcification of
the vessel wall; (L) tortuous distal vessels (black arrow points to the distal Cx mild tortuosity); (M) black arrow points to mid-LAD in diastole
with an obvious myocardial bridge during systole (N); (O) distal LAD fistula to LV chamber (black arrow). Abbreviations: LAD, left anterior
descending artery; RCA, right coronary artery; Cx, circumflex artery.
[2/7/010/8:43:31] [208–228]
Figure 18.5 Optimal view map for each coronary vessel segment. Note that the quadrants represent the angulation and rotation of the
gantry. Abbreviations: LM, left main; LAD, left anterior descending artery; LCX, left circumflex artery; DIAG, diagonal; OM, obtuse marginal;
PDA, posterior descending artery; PL, posterolateral; RCA, right coronary artery; Biff, bifurcation; p, proximal; m, mid; d, distal; RAO, right
anterior oblique; LAO, left anterior oblique; CRAN, cranial; CAUD, caudal.
Figure 18.6 Basic representation of several coronary segment views. Please note the similarities in views when compared with the universal
optimal view map. Abbreviations: RAO, right anterior oblique; LAO, left anterior oblique; LM, left main; LAD, left anterior descending artery;
CX, circumflex, O, obtuse marginal; D, diagonal; RCA, right coronary artery; p, proximal; m, mid. Source: Adapted from Ref. 9.
ability to quantify plaque composition and volume (15). The eccentric stenoses, myocardial bridging and vessel recanali-
proper vessel information is highly dependent on the ability zation are also considered pitfalls of coronary angiography,
of the operator to compare with a normal vessel reference overlap and foreshortening remain the most common and
segment. Even experienced operators are subject to interpre- important imaging limitations associated with the projection
tation limitations and variability (2,19). These limitations have nature of the images (3). Radiation exposure and contrast use
open the door for technologies that evaluate the vessel lumen with the subsequent risk of nephropathy remain significant
via ultrasound or perform pressure differential evaluation of limitations of current X ray–based angiographic practices as
the coronary segments (2). While unrecognized occlusions, well (20–22).
[2/7/010/8:43:31] [208–228]
Vessel Overlap, Foreshortening, and Ostial angiography are generally reversible. Embolization of air is also
Lesions a potential complication but can be easily prevented with
Vessel foreshortening and overlap are recognized imaging newer injectors designed to minimize its risk and also following
artifacts that result from the 2-D angiographic projection of 3- basic but necessary precautions when dealing with a manifold.
D vascular structures (Fig. 18.1). The tortuosity of coronary Nerve damage and or pain have been also described (39). The
artery segments present specific challenges in minimizing ves- widely publicized risk of lactic acidosis with metformin has
sel foreshortening and angiographically separating adjacent been decreased by avoiding the medication before and after the
structures. Historically, the LM coronary artery is difficult to procedure especially among those with renal insufficiency and
image because of foreshortening and vessel overlap from the by monitoring the serum creatinine (40).
circumflex and LAD (6,23,24). Severe allergic reactions are uncommon and easily pre-
The clinical implications of unrecognized foreshortening ventable by the premedication of those at risk 18 to 24 hours
include missed lesions, errors in lesion length assessment, before the procedure. Prednisone 20 to 40 mg, and cimetidine
incomplete coverage of lesions by stents, underestimation of 300 mg every six hours, diphenhydramine, and the use of
stenosis severity, and inaccurate quantitative coronary angiog- nonionic contrast minimize risk. A dose of steroids prior to
raphy calculations. In an era of complex and expensive inter- the coronary angiography has also been successful in decreas-
ventions with drug-eluting stents and other devices, precise ing the risk of a reaction (9). A severe in-laboratory reaction can
length measurements and accurate device placement are criti- be treated with the use of IV epinephrine (0.1 mg ¼ 1 mL of the
cal to minimize the need for additional interventions resulting 1:10,000 solution) every two minutes until the systemic blood
from inaccurate or incomplete imaging. Recognition of these pressure and the wheezing respond favorably.
limitations has resulted in the development of alternative Renal insufficiency is also a potential complication of
imaging techniques designed to specifically address the weak- coronary angiography. Measures like early hydration (pre-
nesses of traditional angiographic techniques. Technologies and postprocedure), minimizing contrast volume and the use
and image processing approaches that are capable of minimiz- of low-osmolar contrast agent are of benefit. Other agents like
ing the shortcomings of traditional angiography have also been N-acetyl-L-cysteine (NAC) and D5/NaHCO3 hydration have
developed and are now in clinical use (8,10,14,25–27). been used to decrease the risk of renal failure in vulnerable
Ostial lesions are traditionally difficult to image because patients with conflicting results (22,28).
of challenges in avoiding the ostium of the vessel to be covered Despite the fact that more complicated and morbidly
by the main branch or a contiguous vessel (Fig. 18.1). Screening compromised patients are being referred for coronary angiog-
angiographic evaluations are often limited in providing a raphy, the rates of complications have not significantly changed
complete ostial vessel survey requiring multiple subsequent (31,34,41). Although radiation safety is the focus of chapter 3, it
angiograms. is worth mentioning that while the advances of angiography
have increased utilization and allowed for patients to fre-
quently come back for further management, this repeated
Complications of Standard Coronary Angiography exposure increases the risk of cumulative radiation injury
The early days of coronary angiography relied on larger diam- (42,43).
eter catheters and high osmolar iodine containing contrast
media. Several adverse effects to coronary angiography have
been described but have significantly decreased over time as SINGLE- AND DUAL-AXIS ROTATIONAL
catheters, techniques and contrast agents have evolved. Once CORONARY ANGIOGRAPHY
the coronary vessel was engaged the replacement of blood with Fundamentals of Rotational Coronary
contrast media resulted in several signs and symptoms. Tran- Angiography
sient hypotension, elevation of left ventricular end diastolic Single-axis RA (Fig. 18.2) is a novel image acquisition technique
pressure, T-wave inversion, sinus slowing, sinus arrest, PR which was designed and developed to address some of the
segment prolongation, QRS prolongation and QT interval pro- limitations of traditional angiography (25,27,44,45). The justifi-
longation were effects seen in response to high osmolar contrast cation for rotational acquisition is straightforward: coronary
media. Arrhythmias that included ventricular tachycardia/ arteries and other vascular structures must be visualized from
ventricular fibrillation, myocardial ischemia due to decreased multiple projection angles to adequately appreciate their struc-
oxygen delivery, allergic reaction and renal toxicity were also ture in the resultant 2-D X-ray projection images (Fig. 18.7). RA
seen (9,22,28–30). Some of these adverse effects have been acquires the same images as traditional angiography but is
reduced with the use of low-osmolar contrast media. Contrast automated, can be standardized, and provides an extensive
agents are the focus of chapter 4. Constant hemodynamic panoramic view of key anatomic features for diagnostic and
monitoring is therefore necessary during coronary angiogra- interventional purposes (Fig. 18.7).
phy. Major complications in contemporary practices are Both standard and RA acquire runs of projections images
uncommon (<1%). They include death (0.10–0.14%), myocar- but the imaging perspective is fundamentally different. Coro-
dial infarction (0.06–0.07%), contrast agent reaction (0.23%), and nary angiography has traditionally employed acquisitions with
local vascular complications (0.24–0.1%) (31–33). Vascular a fixed projection during each injection and the gantry is moved
access and complications are the focus of chapter 8. The risk to multiple positions between additional imaging runs that are
of death is accentuated in vulnerable patients like those with acquired with different degrees of right and LAO rotation and
significant LM disease (0.55%), EF < 30% (0.30%), NYHA different degrees of cranial and caudal angulation.
functional class IV (0.29%), and severe aortic stenosis. Stroke Single-axis RA requires a cranial (Fig. 18.2) and a caudal
is rather uncommon during angiography (3,34) but may acquisition for the LCA and a cranial acquisition for the RCA.
develop because of cholesterol, atherosclerotic, or clot emboli- The preset acquisition arc rotates from the beginning position
zation (35–38). The results of an embolic stroke during routine to the end position during image acquisition producing a set of
[2/7/010/8:43:31] [208–228]
angulating toward RAO cranial and ending at LAO cranial

(Fig. 18.3).
An additional rotational angiographic technique is the
1808 to 2408 acquisitions with rapid sweeps of the gantry in a
single but obviously large arc. A prolonged single-axis 1808
rotations from LAO 1208 to RAO 608 with no angulation has
been used for automatic gated vessel reconstructions providing
volumetric data and thus allowing images to be displayed in
cross-sectional formats similar to IVUS and in maximum inten-
sity projection (MIP) formats similar to that of CTA which will
be discussed later in this chapter (Fig. 18.8) (10).
Therefore, coronary angiography acquisition techniques
have expanded with the advent of new gantry systems that are
capable of performing rapid, standardized, preprogrammed,
and increasingly complex changes in position that deliver
images from hundreds of perspectives. These different angio-
graphic techniques including standard angiography, single-
axis RA, and dual-axis RA are further described on Figure 18.9.
Advantages of Rotational Angiography

The rotational representation of the coronary tree also allows a
better 3-D understanding of the spatial relationships of coro-
nary tree branches by virtue of the rotating 2-D image. RA
protocols can provide up to 360 projections of the arterial tree
Figure 18.7 Single-axis rotational coronary angiography represen- from different angles during a single coronary injection as
tation of trajectory images. Note how with two rotational acquisitions oppose to the limited views provided by standard angiogra-
a complete survey of the anatomy is delivered. The cranial rotation phy. The safety and efficiency of diagnostic RA have been well
goes from RAO 558 to LAO 558. The caudal rotation goes from RAO established with a significant reduction in contrast and radia-
558 to LAO 558. Abbreviations: LAO, left anterior oblique; RAO, right tion exposure while maintaining similar or lower procedural
anterior oblique; CRAN, cranial; CAUD, caudal. times. The data shows that RA has a 30% to 40% reduction on
contrast exposure and at least a 15% reduction in radiation
exposure when compared with standard angiography
(27,44,45). Other studies show a similar contrast exposure
images with a changing perspective that may maximize coro- reduction but fail to show a significant reduction in radiation
nary visualization. A complete study will therefore often have exposure (44). This is in part related to the different protocols
only three total rotational acquisitions each with an RAO to used and the inherent difference in the imaging systems (flat X-
LAO arc of 1108 to 1208 (one cranial for the LCA, one caudal for ray detector vs. an II with or without cardiac optimization).
the LCA, and a cranial for the RCA). Finally, the fact that RA uses a larger field of view is an
Dual-axis RA is a more recent development. Rather than important factor in the reduction of total radiation although
a simple arc, dual-axis rotations may be programmed to move prior studies do show that the reduction is mostly driven by the
the gantry simultaneously in both the RAO-LAO and cranial- acquisition of fewer images (27). While all the studies have
caudal axes. For example, the LCA angiogram has now evolved been with the single-axis RA platform, newer studies with the
into a single acquisition that rotates from LAO caudal to RAO dual-axis RA promise to deliver lower radiation exposure and
caudal subsequently breaking the “X”-axis (dual axis) and total contrast volume use.
Figure 18.8 3-D reconstruction of the left coronary artery based on 1808 rotational angiography.
[2/7/010/8:43:31] [208–228]
Figure 18.9 Representation of the imaging trajectories or sites of standard coronary angiography (circled) rotational coronary angiography
(dotted lines) and DARCA (solid lines) for each coronary vessel. Abbreviations: LAO, left anterior oblique; RAO, right anterior oblique;
DARCA, dual motion rotational coronary angiography.
Last but not least RA provides the perfect platform for validated (6–8,14,16,24,46,47). The accuracy of the 3-D modeling
several advanced imaging applications like on-line 3-D model- method is dependent on a computer-based, four-step algorithm
ing with optimal view map creation (Fig. 18.1), rapid 3-D which integrates 2-D projections into a 3-D image. Modeling is
modeling, and manual or automated gated vessel reconstruc- useful in the sense that it only requires orthogonal views of a
tions (Fig. 18.8). given structure. Given the lack of volumetric data it is not as
precise as a reconstruction, but it allows 3-D imaging of tradi-
tionally difficult to reconstruct structures like the coronary
Limitations of Rotational Angiography arteries. This is mostly important in standard angiography
Despite the visual appeal, better safety and the standardized
laboratories without rotational capabilities. RA also provides
acquisition technique of RA, the inherent limitations of 2-D
images appropriate for 3-D modeling techniques (Table 18.5).
projection imaging remain. Furthermore operators and staff
must isocenter the X-ray system, set up the arc of acquisition
such that anticollision software does not prevent the acquisi-
Coronary Reconstruction
tion, and learn how to review the rotational runs which contain
3-D volumetric data sets can be generated from techniques that
a large amount of visual information. While RA may improve
acquire volumetric data points such as RA, CT, or MR imaging
the ability to visualize a lesion, the accurate selection of the
(Table 18.5). Several methods that are capable of generating 3-D
optimal single projection to show a lesion and perform a PCI is
images have been described and in general, can be divided into
still dependent on the operator’s visual skills. Furthermore, RA
surface- or volume-rendering techniques. The surface-render-
cannot simulate views based on acquired images and provides
ing method relies on a computer algorithm to reconstruct
no quantification of important 3-D vessel features. Hence, the
intensity values which are above a specific defined threshold
need for angiographic 3-D image reconstructions. A list of
and represent volumetric surfaces within the data set; all values
attributes and limitations of all imaging technologies discussed
below the set threshold are discarded and not used for image
in this chapter is provided on Table 18.3.
generation. The resultant image is a representation of the sur-
face contour and appears 3-D through computer-generated
shading. While surface rendering is fast, it uses only a small
3-D CORONARY MODELING AND
portion of the acquired data and is less reliable during imaging
RECONSTRUCTIONS
of structures smaller than 2 to 3 mm. The MIP algorithm is
Two techniques have been developed for the 3-D representation
another commonly used technique. 3-D imaging using volume
of vascular structures including the coronary tree (Table 18.4).
rendering is a more powerful technique that incorporates the
entire data set into the 3-D image. In contrast to surface-
Coronary Modeling rendering techniques, intravascular details and spatial relation-
The 3-D modeling technique uses 3-D centerline data and ships between adjacent structures are preserved (47,48).
shaded or rendered surfaces; the diameter and 3-D morphologic The 3-D “reconstruction” technique is dependent on
structure of the vessel is subsequently derived with a computer multiple image projections for the creation of a volumetric
algorithm (Fig. 18.10). The 3-D “modeling” technique uses two representation of the vessel (Fig. 18.8). The 3-D reconstruction
or more angiographic projections to extract features of the vessel technique refers to a computer-generated representation of the
and create a 3-D representation (Fig. 18.1). A 3-D modeling true shape and size of the imaged vessel using actual volumet-
algorithm using single-plane angiography that does not require ric data obtained from RA, CT (Fig. 18.11), or MR (Fig. 18.8)
a calibration object has been developed and prospectively (44,45,49,50). Interest in improving patient outcomes by
[2/7/010/8:43:31] [208–228]
Table 18.3 Comparison of Coronary Angiographic Imaging Technologies

Invasive angiography Noninvasive angiography
1808 posteroanterior Single- and dual- Computed Magnetic
Standard rotational axis rotational tomography resonance
angiography angiography angiography angiography angiography
Contrast exposure þ/þþ þþ þþþ þ/þþ Gadolinium
Radiation exposure þþ þþþ þþþ þ Not applicable
Procedural time þþ þþ þþþ þþ þ
Invasiveness þ þ þ þþþ þþþ
Image quality þþþ þþ þþþ þþþ þþ
Minimizes imaging þ þþþ þþ þþþ þþþ
Inaccuracies
Provides volumetric data þa þþþ þþb þþþ þþþ
Allows for preprocedural þ þþ þ þþþ þþþ
planning
Applicability in unstable þþþ þþ þþþ þ þ
patients Elevated HR ¼ þ Elevated HR = þþþ
Low BP ¼ þ Low BP = þ
Applicability in obese þþþ þ/þþ þ/þþ þ þ
patients
Compatibility with þþþ þþþ þþþ þ þ
implanted devices
Applicability to calcified þþþ þþþ þþþ þ/þþ þ/þþ
vessels
Applicability to chronic þþþ þþþ þ/þþ þ þ
obstructive pulmonary
disease patients
Note: þþþ, very favorable; þþ, moderately favorable; þ, least favorable.
a
Requires 3-D vessel modeling.
b
Requires 3-D modeling or 3-D/4-D vessel reconstruction.
Abbreviations: 3-D, three-dimensional; 4-D, four-dimensional; HR, heart rate; BP, blood pressure.
Table 18.4 Coronary Angiographic Data Processing Techniques

3-D modeling—3-D representation 3-D reconstruction
Image input >2 orthogonal views (>308) Multiple images required
Image source 3-D centerline based True shape reconstruction
Vessel measurements Size and diameter are derived from Uses actual volumetric data
a computer algorithm
Surface rendering n/a 3-D surface appearance is the product of shading
Volume rendering and maximum n/a Incorporates entire data set in image; Intravascular details,
intensity projection spatial relationship and structures are preserved
Abbreviations: 3-D, three-dimensional; n/a, not applicable.
optimizing the imaging of complex visual structures has been contemporary laboratory has been tested (Figs. 18.1 and 18.10)
paramount in the design and creation of 3-D imaging techni- (12,25,47,51).
ques. Over the last decade, significant progress has been made
in the development of new invasive and noninvasive imaging
techniques that permit accurate, real-time, 3-D displays of
Advantages of 3-D Imaging
Quantifying 3-D vessel properties and characteristics of a given
structures.
vessel are necessary in contemporary interventional cardiology.
The ACC and the AHA outline the specific vessel properties
that define the procedure outcome risk of a given lesion
Coronary Optimal View Map (Table 18.6) (52). It is important to note that these vessel
3-D modeling and 3-D reconstructions are able to provide the characteristics are inherently 3-D in nature therefore a complete
data sets that allow for the development of an optimal view analysis requires a 3-D evaluation. Traditionally, the cardiolo-
map. This optimal view map can be generated for each coro- gist has been unable to quantify some of this important
nary segment and provides foreshortening and overlap variables relying on visual estimation of given characteristics
quantification for all segments for all gantry positions (25). like tortuosity, flexion, torsion and displacement of a vessel.
The use of these 3-D modeling technique and view maps in the Standardization of these vessel characteristics will lead to a
[2/7/010/8:43:31] [208–228]
Figure 18.11 (See color insert) Computed tomography angiogra-

phy rendering with subsequent image reconstruction.
Figure 18.10 (See color insert) Three-dimensional representation
of a “modeled” coronary tree with the subsequent optimal view map
generation of the proximal left anterior descending artery. Please
note the potential angiographic views on the map with the respective
color representing foreshortening. White represents least amount of foreshortening and minimal overlap (12,13,18,25). In addition
foreshortening, whereas red represent the most amount of fore- recent X ray–based advancements have allowed for the direct
shortening. The current LAO 1 CAUD 1 view angiographically has a visualization of in vivo stents with 3-D reconstructions (53).
47% degree of foreshortening visually and on the optimal view map.
Abbreviations: LAO, left anterior oblique; RAO, right anterior obli-
que; CRAN, cranial; CAUD, caudal; Fore., foreshortening; MAGN, LESION CHARACTERISTICS
magnification; Min, minimum; Max, maximum.
There is still controversy as to what diameter stenosis defines
significant CAD. It is well known that typically a 70% lesion
can provoke ischemia (54). The CASS criteria for defining
more comprehensive evaluation of the vasculature that should significant disease was diameter stenosis > 50% in the LM or
subsequently improve procedural outcomes. diameter stenosis > 70% in any other major epicardial vessel or
3-D images via optimal view maps or simple direct branch. An inherent limitation of coronary angiography is in
evaluation can provide information on images with minimal the assessment of stenosis severity due to the problems of
Table 18.5 Comparison Between 3-D Modeling and 3-D Reconstruction

3-D modeling 3-D reconstruction
Ease of use þþþ þþa
Coronary measurements reliability Vessel size þ þþþ
Vessel length þþþ
Image acquisition requirements (number views) þþþ þ (requires rotational angiography)
Standard angiography compatible Yes No
Rotational angiography compatible Yes Yes (requires 1808 rotation)
Computed tomography angiography based n/a Yes
Magnetic resonance angiography based n/a Yes
Note: þþþ, very favorable; þþ, moderately favorable; þ, least favorable.
a
Newer algorithms are now fully automated.
Table 18.6 Type B1 Lesions Have One Characteristic of Type B Lesions, Whereas Type B2 Lesions Have Two or More of the Type B
Lesion Characteristics
Type A Type B Type C
Length <10 mm 10–20 mm >20 mm
Lesion morphology Concentric Eccentric Excessive tortuosity
Pathway to lesion Easy access Moderate tortousity Extreme angulation >908
Angulation <458 >458 but <908 >908
Lesion contour Smooth Irregular -
Calcium in vessel Little or none Moderate to heavy -
% Stenosis Not occluded Occlusion <3 mo -
Location Not ostial Ostial -
SB No major SB True bifurcation lesion Unable to protect SB
Thrombus presence None Some Degenerative vein graft
Abbreviation: SB, side branch.
[2/7/010/8:43:31] [208–228]
Table 18.7 Coronary Angiographic Lesions and Their Respective Description

Angiographic lesion Description
Angulated lesion Angle from the vessel to the proximal lesion
Collaterals (Fig. 18.4) Networks of tiny anastamotic branches interconnect the major coronary arteries
Contrast streaming (Fig. 18.4) Inadequate vessel opacification results in uneven contrast lining of the vessel suggesting
pseudolesions
Coronary aneurysm (Fig. 18.4) Focal dilation of a coronary segment that includes all elements of the vessel wall.
Coronary artery fistula (Fig. 18.4) Communication between a coronary and the pulmonary artery, aorta, a cardiac vein, another
artery or a chamber
Coronary ectasia (Fig. 18.4) Enlargement or increase of the size of a coronary artery
Coronary spasm (Fig. 18.4) Transient vessel contraction (can be catheter induced)
Coronary thrombus (Fig. 18.4) Filling defects consistent with clot within the arterial lumen
Diffuse disease (Fig. 18.4) Long segments with varying degrees of stenosis
Eccentric lesion (Fig. 18.4) Vessel lesion characterized by an uneven distribution of the plaque in relation to vessel wall
Foreshortening (Fig. 18.1D) Imaging phenomenon that occurs on the basis of perpendicularity of the vessel to X-ray beam
Muscle bridge (Fig. 18.4) A dip in the coronary artery below the epicardial surface under small strips of myocardium
Ostial lesion and bifurcation lesion Ostial is a lesion located 3 mm of the origin of the vessel. Bifurcation disease is that o the main
(Fig. 18.1B) branch and the side branch
Total occlusion (Fig. 18.4) A complete lack of flow into a closed segment
Vessel calcification (Fig. 18.4) Presence of calcium in the vessel
Vessel dissection (Fig. 18.4) Creation of a false lumen between the intima and the media
Vessel overlap (Fig. 18.1A) Imaging phenomenon that occurs when vessels or coronary segments are one on top of the other
Vessel tortuosity (Fig. 18.4) Refers to the lack of longitudinal path of the coronary. Frequent turns are of the vessel are seen
defining a normal reference segment when disease is diffuse reconstructions have no added diagnostic value other than
and the eccentric nature of atherosclerosis. Lesion morphology the documentation of the overall findings, a better understand-
and lesion characteristics are equally important when defining ing of 3-D branching patterns, and vessel tortuosity (3).
CAD (Table 18.6). This lesion classification allows the estima- Currently, there is much interest in the assessment of
tion of PCI procedural risk. Type A are simple lesions, type B CAD using MDCT. Recent advances in MDCT have provided
are moderately complex lesions divided into B1 (lesions with the opportunity to noninvasively and three-dimensionally eval-
only one B characteristic), and B2 (lesions with more than three uate the coronary vasculature in a safe and efficient manner.
B characteristics), and C are high risk (52–56). Further angio- Newer CT imaging technology with faster gantry rotations,
graphic lesion characteristics are shown in Figure 18.4 and dual X-ray source scanners, multidetector 64, 128, 256, and
Table 18.7. recently 320 row acquisitions and ECG gating has substantially
improved both temporal and spatial resolutions to adequately
visualize the moving coronary vasculature. Current-generation
MDCT scanners are able to achieve a spatial resolution of
CORONARY COMPUTED TOMOGRAPHY 0.4 mm with a temporal resolution as low as 83 milliseconds
ANGIOGRAPHY during a less than 15-second cardiac acquisition. Initial
Fundamentals relatively small studies evaluating the diagnostic accuracy of
The cardiac applications of CT are a rapidly growing diagnostic 64-slice MDCT compared with diagnostic cardiac catheteriza-
area because of the ability to visualize plaque burden, artery tion have demonstrated sensitivities ranging from 80% to 94%
calcification, and luminal obstruction noninvasively (57,58). and specificities ranging from 95% to 97% (60,61).
Several studies have shown the indications that could aid Routine evaluation of coronary MDCT involves segmen-
physicians in the management of symptomatic and asympto- tation of the individual visualized coronary vessels. From the
matic patients. A symptomatic patient that has no calcification resulting coronary tree, determinations are easily made regard-
is associated with both a lower risk of an abnormal nuclear ing vessel length, curvature, branching angles, stenosis length,
study and angiographic obstruction. The invasive nature, location and severity (Fig. 18.12). Additionally, atherosclerotic
expense, and risk resulting from invasive angiography have plaque composition can be easily assessed. Because of high CT
been instrumental in encouraging the development of new attenuation of calcified lesions, they are differentiated from
diagnostic methods that allow the coronary arteries to be fibrous or lipid-rich lesions. These angiographic features are
visualized noninvasively (57). Electron beam tomography and easily displayed on MDCT-derived 3-D volumetric and ana-
multidetector spiral computed tomography (MDCT) have been tomic representations (62–65).
used in an effort to visualize the coronary arteries after the Imaging of coronary stents with CTA is also possible.
administration of intravenous contrast. Both have high spatial Newer algorithms do allow for the evaluation of post interven-
and temporal resolutions as well as excellent signal-to-noise tion patients although the image quality of the stented segment
ratios, which allows major branches of the coronary tree to be lacks reliability. The evaluation of the in-stent restenosis is
depicted (59). The axial images are then used by the software to therefore possible but limited in certain patients and difficult
generate a volume rendering or reconstruction (Fig. 18.11). to quantify with any precision. The visualization of the patency
Although reconstructions are used for the diagnosis of CAD or occlusion of bypass grafts is also possible but limitations do
because the images are visually appealing, the most reliable exist when accurately and completely evaluating the anasto-
images come from the MPI of the vessels (Fig. 18.12). 3-D mosis sites.
[2/7/010/8:43:31] [208–228]
Figure 18.12 (See color insert) Computed tomography angiography of the LAD (A, B, C, D) and the RCA (E, F, G, H). The initial image on
the left (A) shows a MIP image of the LAD followed by a volume rendering (B) contiguous to it. The same applies for the RCA on the right
panel that has a MIP image of the RCA (E) followed by its respective volume rendering (F). The lower panel shows for each vessel a stretched
MIP image displaying each vessel lumen (D, H) in a longitudinal fashion with its respective “intravascular ultrasound” like cut of the lumen (C,
G). The LAD shows moderate calcification in the proximal segment (A, D) with a moderate to severe obstruction that was also seen on
standard angiography. The RCA has very mild proximal calcification (E, H) followed by a widely patent lumen with no significant disease on
the regular MIP (E) image or the stretched MIP (H). Abbreviations: MIP, maximum intensity projection; LAD, left anterior descending artery;
RCA, right coronary artery.
Indications acquisition, are under the weight limit of the table, and have no
The visualization of plaque is paramount in the diagnosis of significant heart rate variability. For the most part it is expected
CAD. The ability of CT to evaluate plaque has been the product that patients not present with any of the following:
of a significant amount of research. The detection, quantifica-
a. Irregular heart rhythm (e.g., atrial fibrillation/flutter, fre-
tion, and characterization of coronary plaque do play a signif-
quent irregular premature ventricular contractions, or pre-
icant role in risk assessment. Under the assumption that lipid-
mature atrial contractions, and high grade heart block)
rich, thin fibrous cap plaques are at risk of rupture with
b. Obese patients (body mass index > 40 Kg/m2)
subsequent thrombosis more so than fibrotic plaque, research-
c. Renal insufficiency, creatinine greater that 1.8 mg/dL
ers have tried to use CT attenuation values to differentiate
d. Hear rate greater than 70 beats/min refractory to heart
plaque types (66–68). The implications and clinical value of
rate–lowering agents (e.g., a combination of b blocker and
stenting vulnerable plaques has not been established. While
calcium channel blocker)
one of the strongest indications for CT relies on its ability for
e. Metallic interference (e.g., surgical clips, pacemakers, and/
risk stratification based on plaque presence evaluation very few
or defibrillator wires, coils, or tissue expander) (69)
clinical guidelines exist in the United States for this imaging
technology and MR. The appropriate use and indications for For CT angiography patients should be able to
coronary angiography with CT are shown on Table 18.2. It is
based on the ACCF/ACR/SCMR/ASNC/NASCI/SCAI/SIR a. hold still,
2006 appropriateness criteria for “Cardiac computed tomogra- b. follow breathing instructions,
phy and cardiac magnetic resonance imaging: a report of the c. take nitroglycerin (for performing coronary CT angiogra-
American College of Cardiology Foundation Quality Strategic phy only),
Directions Committee Appropriateness Criteria Working d. take iodine in spite of steroid prep for contrast allergy, and
Group, American College of Radiology, Society of Cardiovas- e. lift both arms above the shoulders (69).
cular Computed Tomography, Society for Cardiovascular Mag- Improvements in CTA technology that are now coming
netic Resonance, American Society of Nuclear Cardiology, to market allow for much faster acquisition that lessen the need
North American Society for Cardiac Imaging, Society for Car- for slower heart rates and longer breath holds.
diovascular Angiography and Interventions, and Society of
Interventional Radiology.”
Advantages
Although patients are still exposed to radiation and contrast
Contraindications CTA delivers a noninvasive coronary angiographic evaluation
A list of appropriate/uncertain/inappropriate indications for without the added risk of the more invasive SCA. Furthermore,
CT is shown on Table 18.2. There are no absolute contra- some studies support the use of CTA as a safe and effective
indications to perform CTA other than a patient unwilling to noninvasive imaging modality for defining coronary arterial
consent. Risks, such as radiation exposure and contrast adverse anomalies in an appropriate clinical setting, providing detailed
effects, should be considered. It is expected that patients being 3-D anatomic information that may be difficult to obtain with
evaluated can achieve a breath hold, which is required for the invasive angiography (70).
[2/7/010/8:43:31] [208–228]
The use of CTA in the risk stratification, triage, and Limitations

evaluation of patients continues to be evaluated with very Impaired image quality, due to dense calcifications and multi-
promising results (Table 18.2). ple image artifacts including coronary artery motion and
Having the CTA images available for evaluation in breathing artifacts, limits the clinical utility of noninvasive
planned PCI has now introduced the concept of preplanning. coronary angiography. Because of several reasons as outlined
Operators are able to evaluate a myriad of coronary features above some patients become less than ideal candidates for CTA
that help in preparation for the case. Lesion characteristics as the images may be uninterpretable.
(e.g., occlusion, calcification, side branch), pathway to the Patients may receive a considerable amount of radiation
lesion (e.g., angulation, tortuosity), and even the shape of the from X ray–based studies and this includes CTA. A patient’s
aorta can be evaluated before an actual procedure. Potential exposure (effective dose equivalent) from a routine chest X ray
equipment selection (e.g., stent size, stent length, guide is 0.02 to 0.04 mSv. Although this is equipment and operator
catheter) is now a decision that the operator can make even dependent an average diagnostic catheterization is <8 mSv,
before the procedure starts. The usefulness of this approach is while a coronary intervention is around 15 mSv. Multidetector
formally being evaluated by several groups (71,72). coronary tomography is around 20 mSv with a Thallium
3-D imaging using volume rendering is a powerful tech- perfusion test just above that at 22 to 24 mSv. Radiation dose
nique that incorporates the entire data set into the 3-D image. In is being dramatically reduced with “step-and-shoot” technol-
contrast to surface-rendering techniques, intravascular details ogy. Newer algorithms recently developed will allow for a
and spatial relationships between adjacent structures are pre- significant reduction in radiation exposure.
served (Tables 18.4 and 18.5). The vascular model can be refined The presence of any metallic interference (e.g., surgical
further if data from the heart structures and vessel wall are clips, pacemakers, and/or defibrillator wires, coils, or tissue
available from the primary imaging modality, as with CTA, or expander) is a limitation but not a contraindication to the
after fusion of two modalities, such as IVUS and the 3-D coronary imaging capabilities of CTA (69).
lumen from X-ray image–based reconstruction as an example.
This new multimodality fusion concept that incorporates imag-
ing technologies like CTA- and X ray–based images during a
CORONARY MAGNETIC RESONANCE CORONARY
procedure holds the future of image guidance (Fig. 18.13) (73,74).
ANGIOGRAPHY
Cardiac MR imaging is a rapidly evolving noninvasive imaging
modality that will further advance the goal of providing 3-D
imaging. Cardiac MR has become an established imaging
modality for the assessment of various cardiac disorders
including myocardial viability, infiltrative cardiomyopathies,
congenital heart disease, anomalous coronary arteries, bypass
grafts, cardiac masses, aortic and pericardial diseases (75–79).
Magnetic resonance coronary angiography (MRCA) is a
technique that allows a noninvasive visualization of coronary
arteries. MRCA has gained considerable importance as a non-
invasive method to diagnose coronary artery stenoses and is an
area of active research (80–82). The benefit of MRCA is to not
only visualize the coronary arteries, but also to evaluate cardiac
morphology and function in one setting (78,79).
Current techniques with 3-D navigator MRCA imaging
can obtain a coronary artery data set in approximately 10 to
15 minutes. Analogous to MDCT, from a MRCA volumetric
data set, 3-D vessel features including vessel tortuosity, lesion
lengths, bifurcation angles can be evaluated, quantified and
translated to the interventional cardiologist.
Fundamentals
Since the concept of ionizing radiation has been discussed in
chapters 1 and 3 an abbreviated understanding of MR is
required. The physical interaction required for MR is related
to the atomic nucleus. The frequency of the radio wave absorp-
Figure 18.13 CTA and angiography image fusion. Note the overlay
of the real-time angiographic data (small square) on the CTA data. The
tion depends on the strength of the external magnetic field.
catheter is visualized real time under X-ray guidance with an overlay of Because MR therefore does not interfere with the atomic shell,
the CTA showing the position of an otherwise difficult to cannulate which is responsible for chemical binding, it is fundamentally
saphenous vein graft to the left anterior descending artery. Note the safe, unlike ionizing radiation, which may interact with electro
location of the rest of the coronary anatomy highlighted and labeled on binding, damaging molecules such as DNA. MR only interacts
the CTA image. Abbreviations: CTA, computed tomography angio- with unpaired spin atomic nuclei, which are basically seen in
graphy; LAO, left anterior oblique; RCA, right coronary artery; PL, water and fat (hydrogen-1 abundant tissues) (3). Since this
posterolateral; M1, obtuse marginal one; M2, obtuse marginal two; tissues are abundant in the body images with a high signal-
M3, obtuse marginal three. Source: From Refs. 73 and 74. to-noise ratio can be easily obtained. The nuclei when exposed
to a magnetic field will behave as a magnet. The nuclei precess
[2/7/010/8:43:31] [208–228]
randomly about 1.5-T magnetic field at a resonance frequency does changes (69). In April 2005, the Food and Drug Admin-
of 63 MHz, which is in the radio wave range. When excited by istration approved MR imaging studies immediately after
radio waves the nuclei will rotate away from the direction of implantation of sirolimus and paclitaxel-eluting stents (69).
the main magnetic field and precess in a coordinated manner Although in the past gadolinium based contrast for MR
that causes a net magnetization. In its relaxation form this was used in all patient populations’ appropriate patient selec-
signal is then captured as radio wave echo by the scanner in tion and caution should be exercised when using Gadolinium
a form that can be subsequently transformed into an image by in patients with renal failure due to the risk of nephrogenic
the receiver antenna (3). fibrosing dermopathy (87–89).
The CMR scanner consists of several pieces: (i) the super-
conducting magnet that produces the static magnetic field,
(ii) the radiofrequency amplifier that generates the pulses, Advantages
(iii) the radiofrequency antenna or receiver that captures MRCA without its requirement for more nephrotoxic contrast
them, and (iv) the computer hardware and software that allows agents or exposure to ionizing radiation is an ideal noninvasive
for the management of the data with subsequent image cre- imaging modality to help plan and execute PCI. It allows for
ation. The gating of the images based on the electrocardio- the evaluation of the coronary anatomy, pathway to the lesion,
graphic R-R is what allows for the system to minimize the shape of the aorta, aneurysms, bypass patency, anomalous
movement through the changes of the cardiac cycle and respi- coronary origin, vessel wall characteristics, lesion character-
ratory motion. The acquisition sequence is composed of the istics, coronary flow reserve, and most importantly can couple
following block components: (i) cardiac triggering to suppress that with myocardial viability and potentially function. Because
cardiac motion, (ii) respiratory motion suppression, (iii) pre- of its specificity a normal coronary MRI suggests the absence of
pulses to enhance contrast to noise ratio (CNR) of the coronary severe multivessel disease.
blood, and (iv) image acquisition enhancement. Multiple meth-
ods for coronary evaluation exist and include conventional
spin-echo coronary MRI (very limited), 2-D segmented k Limitations
space gradient echo coronary MRI, 3-D coronary MRI methods Although the safety of MR is well described the full implica-
(the predominant approach for the past decade), the more tions of magnetic field exposure have not been well established
advanced spiral and radial coronary MRI method, and 3-T as it has been for ionizing radiation. It is certainly reasonable
coronary MRI method. that in the modern climate of safety priority, radio wave tech-
Coronary CMR angiography is still technically challeng- nology shares with echocardiography a sizable advantage over
ing and relatively expensive but several studies have shown its ionizing radiation.
value (76,83). Although the imaging limitations of CMR related The challenges for coronary MRA include compensation
to its spatial resolution due challenge its ability to evaluate CAD for cardiac and respiratory motion, spatial resolution and cov-
with confidence, it is very useful in the diagnosis and recognition erage, high level of tortuosity of the coronary vessel, and signal-
of anomalous coronary origins (84). CMR has also been used in to-noise limitations due to adjacent epicardial fat and myocar-
the evaluation of coronary flow (adenosine stress related evalu- dium. Another limitation is that provided by the bare metal
ation) reserve and the subsequent diagnosis of a significant nature of stents (not so much tantulom). MRI has a sizable
coronary lesion and on saphenous vein graft evaluations (85,86). problem when evaluating coronary stents because of signal/
voids artifacts at the site of the stent. Bypass graft imaging is
limited by local signal loss/artifact caused by implanted metal-
Indications
lic objects (hemostatic clips, ostial stainless steel graft markers,
There is no formal guideline for use of MRCA but Table 18.2
sternal wires, prosthetic valves, struts, rings, and graft stents).
does include the appropriateness criteria from the ACCF/
The technical aspects of MRCA are however quickly evolving
ACR/SCMR/ASNC/NASCI/SCAI/SIR 2006 appropriateness
and multiple methods have been used to improve imaging
criteria for “Cardiac computed tomography and cardiac MR
(cardiac and respiratory motion compensation, breath-hold
imaging: a report of the American College of Cardiology Foun-
method, free-breathing methods, MR navigators—triggering
dation Quality Strategic Directions Committee Appropriateness
alone, MR navigators—gating and slice tracking, electrocardio-
Criteria Working Group, American College of Radiology, Soci-
graphic timing, and respiratory suppression methodology).
ety of Cardiovascular Computed Tomography, Society for Car-
Similar to cardiac CTA, ECG triggering is mandatory to prevent
diovascular Magnetic Resonance, American Society of Nuclear
vessel blurring due to intrinsic cardiac motion. To suppress the
Cardiology, North American Society for Cardiac Imaging, Soci-
effects of respiration, MRCA using respiratory gating (naviga-
ety for Cardiovascular Angiography and Interventions, and
tor echo technique) is performed to monitor diaphragmatic
Society of Interventional Radiology.”
motion. Current isotropic fast 3-D techniques provide thinner
slices, superior signal-to-noise ratios and total coverage of
Contraindications coronary arteries over 2-D MRCA techniques. Introduction of
Unlike CTA there are currently several contraindications to new intravascular contrast agents, novel data (k space) sam-
the use of MRCA. Patients are assumed not to present with pling strategies, and higher field strength (3 T) imaging will
severe claustrophobia or specific metallic devices that are further enhance MRCA (90).
contraindicated such as pacemakers, defibrillators, and certain A serous disadvantage of MR is related to its inability to
aneurysm clips. image patients with metallic implants. Although it is safe on
Although studies are ongoing and have evaluated the valve prosthesis, vascular stents and orthopedic implants it has
safety of pacemakers and defibrillators when exposed to MR a serious limitation with pacemakers and defibrillators (pro-
the appropriateness criteria document from ACCF/ACR/ gramming may change). Flying projectiles in the magnetic field
SCMR/ASNC/NASCI/SCAI/SIR 2006 still does not reflect have the potential to strike the patient or a staff member when
[2/7/010/8:43:31] [208–228]
the magnet is activated. Caution to avoid any metallic objects 13. Green NE, Chen SY, Messenger JC, et al. Three-dimensional
should be constantly exercised. vascular angiography. Curr Probl Cardiol 2004; 29:104–142.
14. Messenger JC, Chen SY, Carroll JD, et al. 3D coronary reconstruc-
tion from routine single-plane coronary angiograms: clinical
CONCLUSIONS validation and quantitative analysis of the right coronary artery
Contemporary medicine has a diverse armamentarium of inva- in 100 patients. Int J Card Imaging 2000; 16:413–427.
15. Topol EJ, Nissen SE. Our preoccupation with coronary luminol-
sive and noninvasive imaging technologies to evaluate the cor-
ogy. The dissociation between clinical and angiographic findings
onary anatomy. The risk and benefits of each available imaging in ischemic heart disease. Circulation 1995; 92:2333–2342.
technique should be taken into consideration and tailored to the 16. Gradaus R, Mathies K, Breithardt G, et al. Clinical assessment of a
individual need of the clinical situation. This diversity allows the new real time 3D quantitative coronary angiography system:
clinician the luxury of choosing an imaging evaluation that evaluation in stented vessel segments. Catheter Cardiovasc Interv
minimizes the patient risk while delivering superior results. 2006; 68:44–49.
17. Kern MJ, ed. The Cardiac Catheterization Handbook. 4th ed.
Philadelphia, Pennsylvania: Mosby, 2003.
ACKNOWLEDGMENT 18. Garcia JA. Optimal angiographic views based on 3D reconstructed
The authors wish to acknowledge Gert Schoonenberg for his models. J Am Coll Cardiol 2007; 49(suppl B):296A.
critical review. 19. Galbraith JE, Murphy ML, de Soyza N. Coronary angiogram inter-
John D. Carroll was coinventor of patented 3-D vascular pretation. Interobserver variability. JAMA 1978; 240:2053–2056.
20. Jorgensen NP, Nossen JO, Borch KW, et al. Safety and tolerability
reconstruction and analysis software (patents have been
of iodixanol in healthy volunteers with reference to two mono-
assigned to The University of Chicago and the University of meric X-ray contrast media. Eur J Radiol 1992; 15:252–257.
Colorado), and he received research grants from Philips Med- 21. Mehran R, Aymong ED, Nikolsky E, et al. A simple risk score for
ical Systems. prediction of contrast-induced nephropathy after percutaneous
coronary intervention: development and initial validation. J Am
Coll Cardiol 2004; 44:1393–1399.
REFERENCES 22. Mehran R, Nikolsky E. Contrast-induced nephropathy: definition,
1. Mueller RL, Sanborn TA. The history of interventional cardiology: epidemiology, and patients at risk. Kidney Int Suppl 2006:S11–S15.
cardiac catheterization, angioplasty, and related interventions. 23. Chen SY, Carroll JD. Kinematic and deformation analysis of 4-D
Am Heart J 1995; 129:146–172. coronary arterial trees reconstructed from cine angiograms. IEEE
2. Runge MS, Ohman EM, eds. Netter’s Cardiology. 1st ed. Teter- Trans Med Imaging 2003; 22:710–721.
boro, New Jersey: Icon Learning Systems LLC, 2004. 24. Chen SY, Carroll JD, Messenger JC. Quantitative analysis of
3. Zipes DP, Libby P, Bonow RO, et al., eds. Heart Disease: A reconstructed 3-D coronary arterial tree and intracoronary devi-
Textbook of Cardiovascular Medicine. 7th ed. Philadelphia, ces. IEEE Trans Med Imaging 2002; 21:724–740.
Pennsylvania: Saunders, 2005. 25. Garcia JA, Chen J, Hansgen A, et al. Rotational angiography (RA)
4. Schwartz JN, Kong Y, Hackel DB, et al. Comparison of angio- and three-dimensional imaging (3-DRA): an available clinical tool.
graphic and postmortem findings in patients with coronary artery Int J Cardiovasc Imaging 2007; 23:9–13.
disease. Am J Cardiol 1975; 36:174–178. 26. Leber AW, Becker A, Knez A, et al. Accuracy of 64-slice computed
5. Spears JR, Sandor T, Baim DS, et al. The minimum error in tomography to classify and quantify plaque volumes in the prox-
estimating coronary luminal cross-sectional area from cineangio- imal coronary system: a comparative study using intravascular
graphic diameter measurements. Cathet Cardiovasc Diagn 1983; ultrasound. J Am Coll Cardiol 2006; 47:672–677.
9:119–128. 27. Maddux JT, Wink O, Messenger JC, et al. Randomized study of
6. Chen SJ, Carroll JD. 3-D reconstruction of coronary arterial tree to the safety and clinical utility of rotational angiography versus
optimize angiographic visualization. IEEE Trans Med Imaging standard angiography in the diagnosis of coronary artery disease.
2000; 19:318–336. Catheter Cardiovasc Interv 2004; 62:167–174.
7. Chen SJ, Hoffmann KR, Carroll JD. Three-dimensional reconstruc- 28. Bertrand ME, Esplugas E, Piessens J, et al. Influence of a nonionic,
tion of coronary arterial tree based on biplane angiograms. SPIE iso-osmolar contrast medium (iodixanol) versus an ionic, low-
Med Imaging 1996; 2710:103–114. osmolar contrast medium (ioxaglate) on major adverse cardiac
8. Chen SY, Metz CE. Improved determination of biplane imaging events in patients undergoing percutaneous transluminal coro-
geometry from two projection images and its application to three- nary angioplasty: a multicenter, randomized, double-blind study.
dimensional reconstruction of coronary arterial trees. Med Phys Visipaque in Percutaneous Transluminal Coronary Angioplasty
1997; 24:633–654. [VIP] Trial investigators. Circulation 2000; 101:131–136.
9. Baim DS, Grossman W, eds. Grossman’s Cardiac Catheterization, 29. Nikolsky E, Aymong ED, Dangas G, et al. Radiocontrast nephr-
Angiography, and Intervention. 6th ed. Philadelphia: Lippincott opathy: identifying the high-risk patient and the implications of
Williams & Wilkins, 2000. exacerbating renal function. Rev Cardiovasc Med 2003; 4(suppl 1):
10. Garcia JA, Chen SY, Messenger JC, et al. Initial clinical experience S7–S14.
of selective coronary angiography using one prolonged injection 30. Pedersen HK, Jacobsen EA, Mortensen E, et al. Contrast-medium-
and a 180 degrees rotational trajectory. Catheter Cardiovasc Interv induced ventricular fibrillation: arrhythmogenic mechanisms and
2007; 70:190–196. the role of antiarrhythmic drugs in dogs. Acad Radiol 1995;
11. Scanlon PJ, Faxon DP, Audet AM, et al. ACC/AHA guidelines for 2:1082–1088.
coronary angiography. A report of the American College of Car- 31. Ammann P, Brunner-La Rocca HP, Angehrn W, et al. Procedural
diology/American Heart Association Task Force on practice complications following diagnostic coronary angiography are
guidelines (Committee on Coronary Angiography). Developed related to the operator’s experience and the catheter size. Catheter
in collaboration with the Society for Cardiac Angiography and Cardiovasc Interv 2003; 59:13–18.
Interventions. J Am Coll Cardiol 1999; 33:1756–1824. 32. Samal AK, White CJ. Percutaneous management of access site
12. Green NE, Chen SY, Hansgen AR, et al. Angiographic views used complications. Catheter Cardiovasc Interv 2002; 57:12–23.
for percutaneous coronary interventions: a three-dimensional 33. Witz M, Cohen Y, Lehmann JM. Retroperitoneal haematoma—a
analysis of physician-determined vs. computer-generated views. serious vascular complication of cardiac catheterisation. Eur J
Catheter Cardiovasc Interv 2005; 64:451–459. Vasc Endovasc Surg 1999; 18:364–365.
[2/7/010/8:43:31] [208–228]
34. Segal AZ, Abernethy WB, Palacios IF, et al. Stroke as a complica- 55. Ellis SG, Vandormael MG, Cowley MJ, et al. Coronary morpho-
tion of cardiac catheterization: risk factors and clinical features. logic and clinical determinants of procedural outcome with
Neurology 2001; 56:975–977. angioplasty for multivessel coronary disease. Implications for
35. Blanco VR, Moris C, Barriales V, et al. Retinal cholesterol emboli patient selection. Multivessel Angioplasty Prognosis Study
during diagnostic cardiac catheterization. Catheter Cardiovasc Group. Circulation 1990; 82:1193–1202.
Interv 2000; 51:323–325. 56. Kastrati A, Schomig A, Elezi S, et al. Prognostic value of the
36. Fukumoto Y, Tsutsui H, Tsuchihashi M, et al. The incidence and modified american college of Cardiology/American heart associ-
risk factors of cholesterol embolization syndrome, a complication ation stenosis morphology classification for long-term angio-
of cardiac catheterization: a prospective study. J Am Coll Cardiol graphic and clinical outcome after coronary stent placement.
2003; 42:211–216. Circulation 1999; 100:1285–1290.
37. Hinchey J, Sweeney PJ. Transient cortical blindness after coronary 57. Budoff MJ, Gopal A, Gopalakrishnan D. Cardiac computed
angiography. Lancet 1998; 351:1513–1514. tomography: diagnostic utility and integration in clinical practice.
38. Jackson JL, Meyer GS, Pettit T. Complications from cardiac Clin Cardiol 2006; 29:I4–I14.
catheterization: analysis of a military database. Mil Med 2000; 58. Budoff MJ, Gul K. Computed tomographic cardiovascular imag-
165:298–301. ing. Semin Ultrasound CT MR 2006; 27:32–41.
39. Kuruvilla A, Kuruttukulam G, Francis B. Femoral neuropathy 59. Budoff MJ. Noninvasive coronary angiography using computed
following cardiac catheterization for balloon mitral valvotomy. tomography. Expert Rev Cardiovasc Ther 2005; 3:123–132.
Int J Cardiol 1999; 71:197–198. 60. Budoff MJ. Can non-invasive CT angiography effectively and
40. Heupler FA Jr. Guidelines for performing angiography in patients safely triage patients? Acad Radiol 2007; 14:899–900.
taking metformin. Members of the Laboratory Performance Stan- 61. Budoff MJ, Gopal A, Gul KM, et al. Prevalence of obstructive
dards Committee of the Society for Cardiac Angiography and coronary artery disease in an outpatient cardiac CT angiography
Interventions. Cathet Cardiovasc Diagn 1998; 43:121–123. environment. Int J Cardiol 2008; 129:32–36.
41. Chandrasekar B, Doucet S, Bilodeau L, et al. Complications of 62. de Feyter P, Mollet NR, Cadermartiri F, et al. MS-CT coronary
cardiac catheterization in the current era: a single-center experi- imaging. J Tnterv Cardiol 2003; 16:465–468.
ence. Catheter Cardiovasc Interv 2001; 52:289–295. 63. Mueller D, Maeder A. Robust semi-automated path extraction for
42. Dehen L, Vilmer C, Humiliere C, et al. Chronic radiodermatitis visualising stenosis of the coronary arteries. Comput Med Imag-
following cardiac catheterisation: a report of two cases and a brief ing Graph 2008; 32:463–475.
review of the literature. Heart 1999; 81:308–312. 64. Nieman K, Rensing B, Munne A, et al. Three-dimensional coro-
43. Shechter G, Devernay F, Coste-Maniere E, et al. Three-dimen- nary anatomy in contrast-enhanced multislice computed tomog-
sional motion tracking of coronary arteries in biplane cineangio- raphy. Prev Cardiol 2002; 5:79–83.
grams. IEEE Trans Med Imaging 2003; 22:493–503. 65. Shapiro MD, Nieman K, Nasir K, et al. Utility of cardiovascular
44. Akhtar M, Vakharia KT, Mishell J, et al. Randomized study of the magnetic resonance to predict left ventricular recovery after pri-
safety and clinical utility of rotational vs. standard coronary mary percutaneous coronary intervention for patients presenting
angiography using a flat-panel detector. Catheter Cardiovasc with acute ST-segment elevation myocardial infarction. Am J
Interv 2005; 66:43–49. Cardiol 2007; 100:211–216.
45. Tommasini G, Camerini A, Gatti A, et al. Panoramic coronary 66. Kopp AF, Schroeder S, Baumbach A, et al. Non-invasive charac-
angiography. J Am Coll Cardiol 1998; 31:871–877. terisation of coronary lesion morphology and composition by
46. Gollapudi RR, Valencia R, Lee SS, et al. Utility of three-dimen- multislice CT: first results in comparison with intracoronary ultra-
sional reconstruction of coronary angiography to guide percuta- sound. Eur Radiol 2001; 11:1607–1611.
neous coronary intervention. Catheter Cardiovasc Interv 2007; 67. Schroeder S, Kopp AF, Burgstahler C. Noninvasive plaque imag-
69:479–482. ing using multislice detector spiral computed tomography. Semin
47. Hoffmann KR, Metz CE, Chen Y. Determination of 3D imaging Thromb Hemost 2007; 33:203–209.
geometry and object configurations from two biplane views: an 68. Schuijf JD, Beck T, Burgstahler C, et al. Differences in plaque
enhancement of the Metz-Fencil technique. Med Phys 1995; composition and distribution in stable coronary artery disease
22:1219–1227. versus acute coronary syndromes; non-invasive evaluation with
48. Kapouleas I, Alavi A, Alves WM, et al. Registration of three- multi-slice computed tomography. Acute Card Care 2007; 9:48–53.
dimensional MR and PET images of the human brain without 69. Hendel RC, Patel MR, Kramer CM, et al. ACCF/ACR/SCCT/
markers. Radiology 1991; 181:731–739. SCMR/ASNC/NASCI/SCAI/SIR 2006 appropriateness criteria
49. Budoff MJ, Oudiz RJ, Zalace CP, et al. Intravenous three-dimen- for cardiac computed tomography and cardiac magnetic reso-
sional coronary angiography using contrast enhanced electron nance imaging: a report of the American College of Cardiology
beam computed tomography. Am J Cardiol 1999; 83:840–845. Foundation Quality Strategic Directions Committee Appropriate-
50. Makela T, Pham QC, Clarysse P, et al. A 3-D model-based regis- ness Criteria Working Group, American College of Radiology,
tration approach for the PET, MR and MCG cardiac data fusion. Society of Cardiovascular Computed Tomography, Society for
Med Image Anal 2003; 7:377–389. Cardiovascular Magnetic Resonance, American Society of Nuclear
51. Agostoni P, Biondi-Zoccai G, Van Langenhove G, et al. Compar- Cardiology, North American Society for Cardiac Imaging, Society
ison of assessment of native coronary arteries by standard versus for Cardiovascular Angiography and Interventions, and Society of
three-dimensional coronary angiography. Am J Cardiol 2008; Interventional Radiology. J Am Coll Cardiol 2006; 48:1475–1497.
102:272–279. 70. Budoff MJ, Ahmed V, Gul KM, et al. Coronary anomalies by
52. Krone RJ, Shaw RE, Klein LW, et al. Evaluation of the American cardiac computed tomographic angiography. Clin Cardiol 2006;
College of Cardiology/American Heart Association and the Soci- 29:489–493.
ety for Coronary Angiography and Interventions lesion classifica- 71. Hecht HS. Applications of multislice coronary computed tomo-
tion system in the current “stent era” of coronary interventions graphic angiography to percutaneous coronary intervention: how
(from the ACC-National Cardiovascular Data Registry). Am J did we ever do without it? Catheter Cardiovasc Interv 2008;
Cardiol 2003; 92:389–394. 71:490–503.
53. Movassaghi B, Schaefer D, Grass M, et al. 3D reconstruction of 72. Otsuka M, Sugahara S, Umeda K, et al. Utility of multislice com-
coronary stents in vivo based on motion compensated X-ray angio- puted tomography as a strategic tool for complex percutaneous
grams. Med Image Comput Comput Assist Interv 2006; 9:177–184. coronary intervention. Int J Cardiovasc Imaging 2008; 24:201–210.
54. Kern MJ, ed. SCAI Interventional Cardiology Review Book. 1st ed. 73. Garcia JA, Eng MH, Wink O, et al. Image guidance of percuta-
Philadelphia: Lippincott Williams and Wilkins, 2007. neous coronary and structural heart disease interventions using
[2/7/010/8:43:31] [208–228]
a CT and fluoroscopy integration. Vasc Dis Manage 2007; 4(3): 82. Manning WJ, Nezafat R, Appelbaum E, et al. Coronary magnetic
89–97. Available at: http://vasculardiseasemanagement.com/ resonance imaging. Cardiol Clin 2007; 25:141–170, vi.
article/7252. 83. Kim WY, Spuentrup E, Buecker A, et al. Cardiovascular magnetic
74. Garcia JA, Bhakta S, Kay J, et al. On-line multi-slice computed resonance imaging of coronary atherothrombosis. J Nucl Cardiol
tomography interactive overlay with conventional X-ray: a new 2005; 12:337–344.
and advanced imaging fusion concept. Int J Cardiol 2009; 133: 84. Taylor AM, Thorne SA, Rubens MB, et al. Coronary artery imag-
e101–e105. ing in grown up congenital heart disease: complementary role of
75. de Feyter PJ, Nieman K, van Ooijen P, et al. Non-invasive coro- magnetic resonance and x-ray coronary angiography. Circulation
nary artery imaging with electron beam computed tomography 2000; 101:1670–1678.
and magnetic resonance imaging. Heart 2000; 84:442–448. 85. Langerak SE, Vliegen HW, Jukema JW, et al. Value of magnetic
76. Nieman K, van Geuns RJ, Wielopolski P, et al. Noninvasive resonance imaging for the noninvasive detection of stenosis in
coronary imaging in the new millennium: a comparison of com- coronary artery bypass grafts and recipient coronary arteries.
puted tomography and magnetic resonance techniques. Rev Car- Circulation 2003; 107:1502–1508.
diovasc Med 2002; 3:77–84. 86. Salm LP, Vliegen HW, Langerak SE, et al. Evaluation of saphenous
77. Sakuma H. Cardiac MRI: current status and future prospective. vein coronary artery bypass graft flow by cardiovascular magnetic
Nippon Igaku Hoshasen Gakkai Zasshi 2003; 63:21–25. resonance. J Cardiovasc Magn Reson 2005; 7:631–637.
78. Sakuma H. Cardiac magnetic resonace imaging]. Kyobu Geka 87. Artunc F, Schanz S, Metze D, et al. Nephrogenic systemic fibrosis.
2007; 60:635–641. Dtsch Med Wochenschr 2008; 133(suppl):F1.
79. Sakuma H, Ichikawa Y, Chino S, et al. Detection of coronary artery 88. Kintossou R, D’Incan M, Chauveau D, et al. Nephrogenic fibros-
stenosis with whole-heart coronary magnetic resonance angiog- ing dermopathy treated with extracorporeal photopheresis: role of
raphy. J Am Coll Cardiol 2006; 48:1946–1950. gadolinium?. Ann Dermatol Venereol 2007; 134:667–671.
80. Appelbaum E, Botnar RM, Yeon SB, et al. Coronary magnetic 89. Prchal D, Holmes DT, Levin A. Nephrogenic systemic fibrosis: the
resonance imaging: current state-of-the-art. Coron Artery Dis story unfolds. Kidney Int 2008; 73:1335–1337.
2005; 16:345–353. 90. Nassenstein K, Waltering KU, Kelle S, et al. Magnetic resonance
81. Maintz D, Botnar RM, Fischbach R, et al. Coronary magnetic coronary angiography with Vasovist: in vivo T1 estimation to
resonance angiography for assessment of the stent lumen: a improve image quality of navigator and breath-hold techniques.
phantom study. J Cardiovasc Magn Reson 2002; 4:359–367. Eur Radiol 2008; 18:103–109.
[21/6/010/20:42:23] [229–241]
19
Cardiac catheterization for pediatric patients

Albert P. Rocchini
INTRODUCTION child or young adult with severe developmental disabilities);

Cardiac catheterization is an important diagnostic and thera- and if the child must remain absolutely still during a critical
peutic procedure in pediatric patient with congenital heart phase of an intervention, such as a device or stent placement.
disease. The four commonly used indications for cardiac cath- Not all interventions require general anesthesia, for example,
eterization in the pediatric patient are to make an anatomic conscious sedation works well for most individuals who
diagnosis, to obtain a hemodynamic assessment, to perform a require pulmonary or aortic valvuloplasties and patent ductus
pharmacologic or catheter-based intervention, and to make an arteriosis (PDA) closure and coil occlusion of venous or arterial
electrophysiologic diagnosis and/or perform an electrophysio- collaterals.
logic intervention. With improvements in noninvasive imaging Appropriate conscious sedation protocols will enable the
(echocardiography and magnetic resonance imaging) the use of majority of diagnostic catheterizations to be managed by nurses
cardiac catheterization to make an anatomic diagnosis has trained in both pediatric sedation and the catheterization lab-
significantly reduced. Today, the decision to catheterize a oratory setting.
child with congenital heart disease is based on whether the
anatomic diagnosis by noninvasive methods is incomplete or
inconsistent with the clinical findings. In the preoperative Pharmacologic Agents
patient, the surgeon’s judgment as to the adequacy of the The drugs and doses commonly used to sedate children in the
noninvasive imaging also is a major factor in determining catheterization laboratory are shown in Table 19.1. These agents
whether or not a cardiac catheterization should be performed. have minimal hemodynamic effects in a well-compensated
This chapter deals only with diagnostic cardiac catheter- patient and the main consideration is airway maintenance
ization in the child with congenital cardiac disease. However, and avoidance of respiratory depression (1).
in actual clinical practice the catheterization laboratory requires Choral hydrate is commonly used to sedate infants. The
that the physician recognize unanticipated indications for inter- onset of action is 15 to 30 minutes and duration of action is two
vention that may become evident during the course of a diag- to four hours. About 10% to 20% of children will become
nostic catheterization and perform these interventions when excitable and uncooperative with choral hydrate. It is also
necessary. Therefore, this chapter will also discuss the follow- important to remember that choral hydrate is metabolized
ing areas that are important when performing a cardiac cath- to trichloroethanol and trichloroacetic acid both of which
eterization in a pediatric patient: sedation for the procedure, are pharmacologically active with a long half-life of over
vascular access, hemodynamic assessment, and angiography. 24 hours (2).
Midazolam is a short acting benzodiazepine commonly
used in pediatric catheterization laboratories. If oral midazolam
is used as a premedication, its onset of action is 15 to 30 minutes
CATHETERIZATION LABORATORY SEDATION
with duration of action of two to four hours. In addition, if
FOR THE PEDIATRIC PATIENT
cardiac output and splanchnic perfusion is reduced, hepatic
The approach to premedication and sedation varies widely
metabolism of midazolam will be reduced and the drug will
among institutions. In some institutions all catheterizations are
accumulate. Intravenous midazolam can cause significant hypo-
performed under general anesthesia, and in many others con-
tension in patients with poorly compensated cardiac failure.
scious sedation is performed either supervised by the pediatric
Opioids provide excellent analgesia and commonly used
cardiologist or by a pediatric anesthesiologist. The goals of seda-
opioids in the catheterization laboratory are morphine and
tion are to ensure patient comfort without airway compromise, to
fentanyl. Unlike the synthetic opioid fentanyl, morphine has
promote amnesia, and to facilitate performance of the procedure
sedative properties in addition to providing analgesia. How-
so it may be undertaken in a safe and efficient manner.
ever, fentanyl has a shorter duration of action and generates
much less histamine release and resultant vasodilation and
Indications for General Anesthesia hypotension than morphine. Chest wall rigidity may occur
The decision to use general anesthesia is determined by both with a rapid bolus of fentanyl, although this is an idiosyncratic
patient and procedural factors. In the following circumstances and dose-related reaction (3).
general anesthesia is necessary: airway issues such as having Ketamine is a phencyclidine derivative that effectively
either paralysis of a diaphragm or vocal cord, and/or obstruc- dissociates the thalamic and limbic systems and provides
tive sleep apnea; when hemodynamic compromise is likely to intense analgesia. It provides hemodynamic stability through
occur during an intervention, such as placement of a ventric- sympathomimetic actions resulting from central stimulation
ular septal defect (VSD) device; if the child is very uncooper- and diminished post ganglionic catecholamine uptake that
ative or likely to become very agitated with sedation (e.g., a results in an increase in both heart rate and blood pressure.
[21/6/010/20:42:23] [229–241]
Table 19.1 Drugs and Dosages of Commonly Used Agents for Conscious Sedation or Anesthesia in the Pediatric
Catheterization Laboratory
Drug Route of administration Dose
Analgesics
Morphine IV bolus 0.1–0.2 mg/kg
IV infusion 25–50 mcg/kg/hr
Fentanyl IV bolus 0.5–1 mcg/kg (maximum 4–5 mcg/kg)
IV infusion 2–5 mcg/kg/hr
Ketamine IV bolus 1–2 mg/kg
IM bolus 5–10 mg/kg
Sedatives
Chloral hydrate po 50–80 mg/kg (maximum 1 g)
Midazolam IV 0.1 mg/kg q 10 min
Intranasal 0.1 mL/kg (maximum 1 cc)
Benadryl po 1 mg/kg
IV 1 mg/kg drip over 20 min
Anesthetics
Propofol IV bolus 1.5–3 mg/kg
IV infusion 50–150 mcg/kg/min
Etomidate IV bolus 0.2–0.3 mg/kg
Ketamine will dilate cerebral vessels and should be avoided in catheters are in place when the infant arrives in the cardiac
patients with intracranial hypertension. There are conflicting catheterization laboratory, they are prepared with a chlorhex-
reports about the effect of ketamine on pulmonary vascular idine gluconate solution, cut near the skin, and exchanged over a
resistance. On balance, it appears that the increase in pulmo- guidewire for the appropriate catheter or sheath. For the umbil-
nary artery pressure with ketamine is minimal and should not ical vein, a 5- or 6-Fr sheath—long enough to cross the ductus
prevent its use in individuals with pulmonary hypertension (4). venosus—is positioned in the inferior vena cava or right atrium.
Although respiratory depression with ketamine can occur, For the umbilical artery, no sheath is necessary; usually a cath-
children continue to breathe spontaneously; however, airway eter can be advanced over the wire into the aorta.
secretions are increased and aspiration and or laryngospasm If catheters have not been placed in the umbilicus, umbil-
can result. The major side effects of ketamine are delirium, ical tape is tied to the base of the umbilicus, the umbilical cord
hallucinations, and nightmares; however, these can be mini- is cut horizontally near the skin, and the artery and vein are
mized by pretreatment with benzodiazepines. cannulated with umbilical catheters. Once the umbilical cathe-
Propofol is a phenol derivative supplied in a soy emulsion ters are positioned in the aorta and/or right atrium they are
and egg phospholipid to make an injectable emulsion. In most exchanged over guidewires as previously described. If the
institutions it is only used by anesthesiologists. It has a short umbilical venous catheter cannot be advanced into the right
duration of action and rapid clearance and is administered by atrium, one can use fluoroscopic guidance and small hand
infusion or repeated bolus doses. Side effects include pain on injections of contrast to demonstrate patency of the ductus
injection and hypotension secondary to a decrease in systemic venous. A torque-controlled wire can then be manipulated
vascular resistance and direct myocardial depression. The into the right atrium and the catheter can be advanced over
hypotension and the myocardial depression have limited its the wire, through the ductus venous into the inferior vena cava
used as the sole anesthetic agent during cardiac catheterization; and right atrium.
however, the combination of propofol and ketamine produces A disadvantage of using the umbilical vessels is that the
hemodynamic stability and excellent analgesia (5). umbilical vein directs the catheter posteriorly in the right
Sevoflurane is the most commonly used inhalation agent. atrium toward the foramen ovale and left atrium; therefore it
Sevoflurane has a rapid onset and a relatively safe hemody- is ideal for advancing the catheter into the left atrium and
namic profile. Sevoflurane can cause hypotension secondary to ventricle and undesirable for advancing the catheter into the
direct myocardial depression, and bradycardia and atrioven- right ventricle and pulmonary artery. Another disadvantage is
tricular conduction blockade have also been reported. that the umbilical arteries enter the internal iliac artery and add
an additional curve to the catheter that makes it difficult to
maneuver.
VASULAR ACCESS
Vascular access can be one of the more difficult parts of a Femoral Access
pediatric cardiac catheterization. Access can be obtained from The Seldinger technique is used to obtain percutaneous entry
the femoral or umbilical vessels, from the subclavian, internal into the femoral vessels in the vast majority of cases (6,7). The
jugular or hepatic veins, from the carotid arteries, or even via child should be positioned with the hips elevated and the leg
transthoracic puncture. should be straightened with slight outward rotation. The land-
marks for determining site of vessel entry are the anterior
superior iliac spine, the pubic tubercle, the inguinal ligament,
Umbilical Access and the femoral pulse. The vessel should be entered below the
For infants the umbical artery and vein are frequently used to inguinal ligament to ensure the ability to obtain hemostasis at
perform diagnostic and therapeutic catheterizations. If umbilical the end of the procedure. The area is prepped with a
[21/6/010/20:42:23] [229–241]
CARDIAC CATHETERIZATION FOR PEDIATRIC PATIENTS 231
chlorhexidine gluconate solution and the skin and subcutane- location. Complications include pneumothorax, hemothorax,
ous tissue anesthetized with lidocaine. Venous access is usually and subclavian artery or aortic puncture.
obtained first but this is not mandatory. In infants, the vessel
should be entered about 1 cm below the inguinal ligament. The
vein is medial to the artery and will be quite close, within 2 mm
Internal Jugular Access
The internal jugular approach is frequently used for right heart
of the artery, in infants; whereas, in older patients it may be a
endomyocardial biopsies, to access the pulmonary arteries in
centimeter away from the artery. The angle between the needle
children with a cavopulmonary connection, and when the
and the skin should be 458 or less. The needle should be
femoral vessels are occluded. An advantage of the internal
advanced almost to bone and then slowly withdrawn with or
jugular approach over the subclavian approach is that the
without gentle aspiration with a syringe. Once blood is seen,
vessel is entered well outside the thorax making pneumothorax
the needle is stabilized and the soft end of a wire is advanced
an unlikely complication. As with subclavian access, the inter-
into the femoral vein. If any resistance is felt while advancing
nal jugular approach is not well suited to cross a patent fora-
the wire, the wire should be withdrawn and the needle read-
men ovale or perform a trans-septal puncture. The other
justed. After using fluoroscopy to confirm that the wire is in the
disadvantage in an infant with the internal jugular approach
correct vessel, a small skin incision is made, the needle is
is that it is difficult to immobilize uncooperative patients and
removed, and a sheath and dilator are advanced over the
more sedation or general anesthesia may be required.
wire. If the wire cannot be advanced despite having excellent
The right internal jugular is preferred as it offers a more
blood return, contrast should be injected into the vein to ensure
direct route to the right atrium. The patient is positioned with
that the common iliac vein is not occluded. A plexus of collat-
the neck hyperextended by placing a roll under the shoulders
eral will be seen if the vessel is occluded or stenotic. If
and the head turned to the opposite side. Always use ultra-
collaterals fill the contralateral iliac vein it is likely that the
sound guidance to enter the internal jugular vein. The vein is
contralateral femoral vein is patent. A similar technique is used
imaged in short axis and the needle can be visualized as it
for obtaining femoral artery access.
enters the vein. Once the needle appears to be in the vein and is
The most common complications of femoral access are
confirmed by free return of blood into a syringe, the soft end of
hematomas, arteriovenous fistulae, pseudoaneurysms, retro-
a guidewire is passed through the needle and entry into the
peritoneal hemorrhages, venous thrombosis, and loss of the
right atrium or pulmonary is confirmed by fluoroscopy. A
arterial pulse. Absence of the arterial pulse is not uncommon in
hemostatic sheath should then be positioned into the vein.
young infants immediately following removal of the arterial
Complications include hemothorax, pneumothorax, carotid
sheath. If the pulse has not returned within 2 hours the infant
artery puncture, and tracheal puncture.
should be heparinized with 100 units/kg, followed by an
infusion of 20 units/kg/hr for up to 12 to 24 hours or until
the pulse returns. If the pulse is still absent at 24 hours, Hepatic Access
streptokinase or tissue plasminogen activator can be started Percutaneous transhepatic venous access is an excellent alter-
(8,9). Surgical thrombectomy may be required in rare cases (10). native to the femoral vein, especially in cases where it is
A pseudoaneurysm of the femoral artery can frequently be necessary to cross an atrial septal defect or perform a trans-
treated with ultrasound-guided compression and thrombin septal puncture (12).
injection (11). A Chiba needle is introduced into the skin at the costal
margin near the anterior axillary line. The needle is advanced
cephalad and posteriorly toward the intrahepatic inferior vena
Subclavian Access cava. The needle is usually advanced to within a few centi-
The subclavian vein is routinely used in individuals with a meters of the right border of the spine and the obturator is
Glenn anastomosis, after a hemi-Fontan procedure, or if the removed and a syringe with contrast is attached. As the needle
femoral vessels are occluded. The major disadvantage with is slowly withdrawn contrast is simultaneously injected until
subclavian access is difficulty in crossing a patent foramen contrast is observed to freely fill the hepatic vein. The syringe is
ovale and the inability to perform a trans-septal puncture. removed and a guidewire inserted. The desired sheath is then
The patient is positioned with the arm down at the side, advanced over the guidewire into the hepatic vein. At the end
with a small roll under the spine, and the head turned to the of the case, hemostasis is achieved with a Gianturco coil placed
opposite side. The landmarks are the depression in the lateral in the hepatic tract. This is accomplished by placing the dilator
third of the clavicle and the suprasternal notch. The skin and of the sheath into the hepatic vein. The dilator and sheath are
subcutaneous tissues and clavicular periosteum are infiltrated slowly withdrawn while injecting contrast into the dilator.
with lidocaine. A needle with a syringe attached is inserted at Once it is determined that the dilator is out of the hepatic
the junction of the medial and middle third of the clavicle (at vein and into the tract an appropriately sized coil is placed in
the depression) and is advanced gently until it contacts the the tract.
clavicle. It is then advanced under the clavicle and then Abdominal pain is frequent following transhepatic access
advanced parallel to the floor and toward the suprasternal but a significant peritoneal hematoma is rare (12).
notch. Care must be taken to avoid passing the needle through
the periosteum as this will make it nearly impossible to intro-
duce the sheath. Once blood is freely being aspirated from the Percutaneous Transthoracic Puncture
needle, the syringe can be removed and the soft end of a wire Percutaneous transthoracic puncture can be used to obtain
advanced under fluoroscopic guidance into the right atrium or access to the ventricles in a patient with a mechanical valve
cavopulmonary anastomosis. If access is not obtained after in the aortic and mitral, or tricuspid position (13). It can also be
several attempts, contrast injection through an IV in the hand used to puncture a surgically isolated left atrium in individuals
or arm should be performed to document vessel patency and following the Fontan procedure (14).
[21/6/010/20:42:23] [229–241]
The procedure should be performed under general anes- vascular resistance, intracardiac pressure gradients, and valve
thesia and requires either transesophageal or transthoracic areas.
echocardiographic guidance. A modified triaxial system can
be used with an 18-gauge lumbar needle inserted in a dilator/ Pressure Measurement
sheath combination. The subxiphoid position is used for right In both adult and pediatric catheterization laboratories, pres-
ventricular (RV) puncture and left atrial puncture after the sures are usually measured with fluid-filled catheters. Six
Fontan procedure, whereas an apical position is used for left common errors can lead to inaccurate measurement of pres-
ventricular puncture. Initially, a 4- to 6-Fr sheath is placed for sure.
diagnostic hemodynamics and angiography and this sheath can
be up-sized to facilitate an intervention. Following completion 1. Air in the system which usually leads to a damped (arti-
of the procedure the sheath is removed and a purse-string ficially lower) pressure; however, occasionally air results in
suture is placed in the superficial skin wound. amplification of the pressure wave leading to overshoot of
Complications can include pericardial effusion and fling (artificially high pressure) (Fig. 19.1).
hemothorax. The procedure should only be done in individuals 2. Loose connection in the system usually results in over-
who have had previous cardiac surgery, since a surgically- damping of the wave form.
scarred pericardium is important for hemostasis especially after 3. Partial catheter obstruction leading to a damped pressure
a ventricular puncture. tracing (this is especially common when using small thin
walled catheters).
4. Catheter fling resulting from the catheter being in a tur-
HEMODYNAMICS bulent jet or if the catheter is struck by a cardiac structure
Four important areas in the hemodynamic evaluation of an such as the anterior leaflet of the mitral valve, will result in
individual with congenital heart disease include measurement an artificially high pressure.
of pressures, measurement of blood oxygen content, measure- 5. Inaccurate calibration can result from either movement of
ment of cardiac output, and measurement of shunt size, the patient and/or the transducer during the study.
Figure 19.1 The effect of an air bubble in the pressure transducer. (A) The left ventricular pressure in a 10-month-old with valvar aortic
stenosis. This pressure was recorded with a small bubble in the transducer. The arrows depict the overshoot (“fling”) produced by the air
bubble. If one used this pressure tracing to measure LV pressure the systolic pressure of 177 mmHg would have markedly over estimated the
pressure. (B) The LV pressure after the air bubble was removed from the transducer. Notice the now normal LV pressure wave form without
the fling. The true LV pressure is 150 mmHg not 177 mmHg. Abbreviation: LV, left ventricle.
[21/6/010/20:42:23] [229–241]
Table 19.2 Normal Hemodynamics for a Child Measurement of Cardiac Output

The most common methods for measuring cardiac output use
Cardiac chamber Normal pressures
the indicator dilution technique first described by Fick (16). The
Right atrium Mean pressure of 3–5 mmHg indicators most commonly used in congenital heart disease are
Right ventricle 20–25/3–5 mmHg oxygen and cold saline (thermodilution) (17). The basic princi-
Pulmonary artery Mean pressure of 13–15 mmHg ples of the indicator dilutions method are that an indicator is
Pulmonary artery wedge Mean pressure of 8–10 mmHg
present in the fluid in a measurable concentration, and the
Left ventricle 96–100/5–10 mmHg
indicator is added or removed at a known rate. If, for example,
Systemic artery 90–110/60–70 mmHg
oxygen is the indicator, the equation to measure cardiac output
is as follows:
6. Catheter entrapment occurs when an end-hole catheter is VO2
Qs ¼
placed in a small or heavily trabeculated chamber and ðAO O2 content MV O2 contentÞ
traps a small volume of fluid that results in an exaggerated VO2
systolic pressure elevation. Qp ¼
ðPV O2 content PA O2 contentÞ
Table 19.2 summarizes the normal hemodynamics for a where Qs is systemic blood flow, AO aortic, MV mixed venous,
child. PV pulmonary venous, PA pulmonary artery, Qp pulmonary
blood flow, and VO2 oxygen consumption.
Blood Oxygen Measurements Using oxygen as the indicator, the most difficult param-
Oxygen is carried in the blood either dissolved in plasma or eter to measure is oxygen consumption. This is either measured
attached to hemoglobin. The amount of oxygen dissolved in the directly using a flow-through method (18) or using an assumed
plasma at 378C is about 0.03 mL/mmHg/L (or 3 mL of value based on the formulas of Lafarge and Miettinen (19).
dissolved oxygen per liter, for every 100 mmHg partial pres- In the thermodilution method, the indicator is tempera-
sure of oxygen). This amount of oxygen is quite small in ture. A double lumen catheter is used and saline is injected in
comparison with the amount of oxygen bound to hemoglobin the proximal port usually in the right atrium and the thermistor
and is therefore usually ignored. However, if the child is in for measuring temperature is positions on the distal end of the
supplemental oxygen, with a PO2 > 100 mmHg, then dissolved catheter usually in the pulmonary artery.
oxygen must be considered in the calculation of total oxygen The thermal dilution method is inaccurate in the follow-
content. ing circumstances: significant pulmonary or tricuspid regurgi-
Most of the oxygen in blood is bound to hemoglobin. The tation, the presence of significant intracardiac shunts, and if the
amount is dependent on the type of hemoglobin, temperature, baseline temperature is unstable (i.e., patient has a fever or at
partial pressure of oxygen and carbon dioxide, and the level of the end of strenuous exercise).
2,3-DPG. The maximum amount of oxygen that can be taken up
by hemoglobin in blood is referred to as the oxygen capacity.
Oxygen capacity can be directly measured by the method of
Measurement of Shunt Size, Vascular Resistance,
Van Slyke (15); however in current practice it is assumed that
Valve Areas, and Intracardiac Pressures
the maximal oxygen capacity is 1.36 mL/g of hemoglobin; thus,
Shunt Detection and Quantification
O2 capacity ðmL=LÞ ¼ Hb ðg=dLÞ 1:36ðmL=gÞ 10 dL=L An increase in oxygen saturation between different sites on the
right side of the heart is used to detect the presence of left-to-
The oxygen content is the amount of oxygen present in a right shunt; whereas a decrease in saturations between differ-
sample of blood, and this includes both dissolved oxygen and ent sites on the left side of the heart is used to detect the
oxygen bound to hemoglobin. presence of a right-to-left shunt. Since the oxygen saturations of
O2 content ðmL=LÞ ¼ ðO2 capacityÞ ðO2 saturationÞ blood in the superior vena cava, inferior vena cava, right
þ ð0:03 mL=mmHg=LÞ ðPO2 mmHgÞ atrium, right ventricle, and pulmonary artery are not the
same even in individuals with no intracardiac shunting, it is
When the child is breathing room air, this can be sim- important to be able to define the minimum change in blood
plified to oxygen saturation that is needed to reliably demonstrate an
O2 content ðmL=LÞ ¼ ðO2 capacityÞ ðO2 saturationÞ intracardiac shunt. In adults it has been reported that an
interchamber oxygen saturation difference of as small as 3%
Oxygen saturation is usually measured using the spec- (right ventricle to pulmonary artery) or as large as 9% (superior
trophotometric method. Measuring oxygen saturation directly vena cava to pulmonary artery) can reliable detect a left-to-right
rather than calculating it from the oxygen-hemoglobin dissoci- shunt (20). Table 19.3 lists the minimum saturation difference to
ation curve makes the measurement independent of factors that detect a shunt at the 99% confidence limit for children (21). If
may increase (alkalosis, hypothermia, fetal hemoglobin) or more than one set of saturations is obtained the minimum
decrease (acidosis, fever) hemoglobin/oxygen affinity. saturation difference is reduced (22).
Table 19.3 Minimum Saturation Difference to Detect a Shunt at the 99% Confidence Limit
Chambers sampled Minimum saturation difference (%) Multiple samples difference (%)
Superior vena cava–right atrium 8.7 7
Right atrium–right ventricle 5.2 4
Right ventricle–pulmonary artery 5.6 4
[21/6/010/20:42:23] [229–241]
If the aortic saturation is <94% or there is a 2% or greater The calculation of vascular resistance is based in part on
step down from left atrium or left ventricle (LV) to aorta then a Poiseuille’s law, which states that flow in a tube is directly
right-to-left shunt is suspected. related to pressure and cross-sectional area of the tube and
The Fick principle is used to calculate pulmonary and inversely to the length of the tube and the viscosity of the fluid
system blood flows. To calculate right-to-left and left-to-right flowing in the tube. In the vascular system the length of the
shunts, the concept of effective pulmonary blood flow must be tube and the viscosity of the fluid are assumed to be constant so
used. Effective pulmonary blood flow (Qep) is defined as the that the pressure gradient across a vascular bed divided by the
desaturated blood that flows to the lungs. In the absence of a flow through the bed is equal to the resistance of the bed.
right-to-left shunt, effective blood flow equals pulmonary Therefore, the formulas for pulmonary and systemic vascular
blood flow. The equation to calculate effective pulmonary resistance are as follows:
blood flow is
ðPA mean pressure LA mean pressureÞ
VO2 Rp ¼
Qep ¼ Qp
ðPV O2 content MV O2 contentÞ
ðAO mean pressure RA mean pressureÞ
Rs ¼
When intracardiac shunts are present, the superior vena Qs
cava is probably the best estimate of mixed venous blood. The
three situations where this may not be the case are individuals There are two different types of pulmonary resistance:
under general anesthesia, the presence of supracardiac partial arteriolar
h resistance is resistance calculated
i across the vascular
or total anomalous pulmonary venous return, and in many ðPA mean pressure LA mean pressureÞ
bed Qp and total resistance of the
individuals with an atrial septal defect in which atrial blood
mean PA pressure
refluxes into the proximal portion of the superior vena cava. lungs Qp . In the clinical setting we are usually
The volume of left-to-right shunt is equal to the differ- only interested in arteriolar resistance.
ence between pulmonary blood flow and effective blood flow. The units for this expression of vascular resistance are
called Wood’s units (named after Paul Wood) and are in
Ql r ¼ Qp Qep
mmHg/L/min. To convert this to metric units, the Wood’s
The volume of a right-to-left shunt is equal to the differ- unit is multiplied by 80 to yield the units of dyneseccm5.
ence between systemic blood flow and effective blood flow. Because of the considerable size range of pediatric patients
most cardiologists index the Wood’s units to body surface area
Qr l ¼ Qs Qep (Wood’s unit m2). The normal values for systemic and pulmo-
nary resistance are <20 units for systemic and <3 Wood’s units
The important exception to this definition is in an infant
for pulmonary.
with D-transposition of the great arteries who has parallel
circulations. In these infants the Qep is the amount of blood
that is mixing between the pulmonary and systemic circuits (i.e., Pulmonary Vasodilator Testing
the left-to-right and right-to-left shunting). In patients with pulmonary hypertension in whom it is impor-
Q
The ratio of pulmonary to systemic blood flow Qps is a tant to determine if their hypertension is fixed or reactive,
useful estimate of the magnitude of left-to-right shunting. Small pulmonary vasodilator testing is performed at the time of a
Qp
left-to-right shunts are defined as Qs of <1.5, moderate 1.5 to 2, diagnostic catheterization.
Qp The two pulmonary vasodilators that are used are oxygen
and large >2. The formula to calculated Qs is and nitric oxide (23,24). The usual protocol is to initially obtain
Qp ðMV saturation AO saturationÞ pulmonary artery, pulmonary venous wedge and aortic pres-
¼ sures, saturations and blood gases, along with a measurement
Qs ðPV saturation PA saturationÞ
of pulmonary blood flow (either with using the Fick procedure
If the patient is in oxygen, then dissolved oxygen must or with thermodilution if other cardiac lesions are not present).
also be taken into account. The formula then becomes Following these baseline measurements, the individual is
placed in 100% oxygen for 10 minutes and repeat measure-
Qp ðMV O2 content AO O2 contentÞ ments are made. If oxygen does not result in normalization of
¼
Qs ðPV O2 content PA O2 contentÞ pulmonary vascular resistance then the individual receives
oxygen along with nitric oxide either at 50 ppm through a
nasal cannula or at 20 ppm though an endotracheal tube. After
Vascular Resistance
10 minutes of the nitric oxide, repeat hemodynamic measure-
One of the most common reasons for performing a diagnostic
ments are made. Combination testing with NO + O2 provides
cardiac catheterization is to assess pulmonary vascular resis-
additional pulmonary vasodilation in patients with a reactive
tance. This is especially important in the following situations:
pulmonary vascular bed in a selective, safe, and expeditious
l To determine if an individual can have their cardiac defect fashion during cardiac catheterization (25). Pulmonary vaso-
repaired, that is, child with a VSD device and pulmonary reactivity can also be assessed by using aerosolized iloprost (a
artery hypertension (Table 19.4) stable carbacyclin derivative of prostacyclin, intravenous epo-
l To determine if the child is a candidate for cardiac trans- prostenol, and/or intravenous adenosine) (26). These agents
plantation are comparable to inhaled NO in their ability to induce pul-
l To determine the hemodynamic response of a child with monary vasodilatation and since they do not require a special
pulmonary artery hypertension who is/or may have to be delivery system they may even be more cost effective. A reac-
treated with pulmonary artery vasodilators tive pulmonary bed is defined by a drop in pulmonary vascular
[21/6/010/20:42:23] [229–241]
Table 19.4 Example of Pulmonary Vasodilator Testing in a 12-Month-Old Infant with a Large Ventricular Septal Defect
Sitea Saturation Pressure PO2
SVC 60% Mean ¼ 5 mmHg 42 mmHg
PA 65% 88/50 (63) mmHg 47 mmHg
LA/PV 97% Mean ¼ 12 mmHg 88 mmHg
Aorta 93% 88/55 (66) mmHg 69 mmHg
Siteb Saturation Pressure PO2
PA 92% 90/30 (49) mmHg 68 mmHg
Aorta 100% 90/60 (70) mmHg 425 mmHg
Sitec Saturation Pressure PO2
PA 99% 88/25 (46) mmHg 108 mmHg
Aorta 100% 88/65 (72) mmHg 400 mmHg
Note: In this child, the combination of O2 and NO normalized Rp by increasing pulmonary blood flow while only slightly
decreasing pulmonary artery pressure.
a
The following hemodynamic measurements were made on diagnostic cardiac catheterization.
Hb ¼ 13 g/dL and VO2 measured at 188 mL/min m2.
188 mL=min m2
Qp ¼ ¼ 3:3 L=min m2
ð13 1:36 10Þ ð:97 65Þ
188 mL=min m2
Qs ¼ ¼ 3:2 L=min m2
ð13 1:36 10Þ ð:93 60Þ
188 mL=min m2
Qef ¼ ¼ 2:9 L=min m2
ð13 1:36 10Þ ð:97 60Þ
Qp
¼ 1:03
Qs
ð63 12Þ
Rp ¼ ¼ 15:4 Wood0 s units m2
3:3
b
The following hemodynamics were present after the child was in 100% O2 for 10 minutes.
188 mL=min m2
Qp ¼ ¼ 7:5 L=min m2
½ð13 1:36 10 1Þ
þ ð0:03 425Þ ½ð13 1:36 10 :92Þ þ ð0:03 68Þ
188 mL=min m2
Qs ¼ ¼ 2:9 L=min m2
½ð13 1:36 10 1Þ þ ð0:03 425Þ ½ð13 1:36 10 :70Þ þ ð0:03 45Þ

Qp
¼ 2:6; Qef ¼ Qs
Qs
ð49 17Þ
7:5
c
The following hemodynamics were present after the child was on O2 and 50-ppm NO for 10 minutes.
188 mL=min m2
Qp ¼ ¼ 17 L=min m2
½ð13 1:36 10 1Þ
þ ð0:03 400Þ ½ð13 1:36 10 :99Þ þ ð0:03 108Þ
188 mL=min m2
Qs ¼ ¼ 2:9 L=min m2
½ð13 1:36 10 1Þ þ ð0:03 400Þ ½ð13 1:36 10 :70Þ þ ð0:03 45Þ
Qp
¼ 5:9; Qef ¼ Qs
Qs
ð46 17Þ
17
Abbreviations: PA, pulmonary artery; LA, left atrium; PV, pulmonary vein; SVC, superior vena cava.
resistance. This can be due to an increase in pulmonary blood pulmonary blood flow (i.e., the child with idiopathic pulmo-
flow with no or little change in pulmonary artery pressure (i.e., nary artery hypertension), or a combination of both. An exam-
the child with a large VSD and pulmonary hypertension), a ple of the calculations association with pulmonary vasodilatory
decrease in pulmonary artery pressure without a change in testing is depicted in Table 19.4.
[21/6/010/20:42:23] [229–241]
Figure 19.2 Right atrial pressure tracing in a child with a large ASD. (A) The right atrial pressure tracing demonstrates equal “a” and “v”
waves. This tracing is characteristic of the atrial pressure tracing in a patient with a large ASD. (B) The right atrial pressure tracing in the same
child after the ASD has been closed with a 22-mm Amplatzer ASD occluder. After the ASD has been closed the right atrial pressure tracing
has normalized with a much larger a wave than v wave. Abbreviation: RA, right atrium; ASD, atrial septal defect.
Valve Areas arterial pressure (>10 mmHg) with inspiration. In individuals

Valves areas are infrequently used in the management of chil- with chronic pericardial constriction, a fluid bolus (10–20 cc/kg
dren with congenital heart disease. Most decisions as to of warm saline) uniformly increases all pressures; whereas, if
whether or not to perform surgery or balloon angioplasty are the individual has chronic myocardial constriction the fluid
based on peak-to-peak pressure gradients at the time of a heart bolus will increase pressure to a greater degree in the cardiac
catheterization. When valve areas are calculated, either the chamber most severely affected (29).
Gorlin formula (27) or Bache’s modification of the Gorlin for- Pressure pullbacks with end-hole catheters are very use-
mula (28) is used. ful in identify the precise site of obstruction, that is, a patient
In addition to measuring absolute systolic, diastolic, and with multiple sites of left ventricular outflow tract (LVOT)
mean pressures, analysis of the intracardiac pressure wave obstruction.
forms and pressure gradients between chambers are very
important in performing a diagnostic cardiac catheterization
in a patient with congenital heart disease. For example, the ANGIOGRAPHY
dominant pressure wave form in the right atrium is the “a” Angiography still remains an essential component of the diag-
wave, whereas the “v” wave is the dominant pressure wave in nostic evaluation of the patient with congenital heart disease.
the left atrium; however with an atrial septal defect the a and v The selection of the appropriate angiographic projection is
are nearly the same in both atrium (Fig. 19.2). critical for obtaining diagnostic images (30,31). The standard
The pressure wave forms are critical for making the angiographic projections currently used in individuals with
diagnosis of both cardiac tamponade and pericardial constric- congenital heart disease are frontal (posteroanterior), lateral,
tion. In tamponade, the pericardial fluid causes equalization of right anterior oblique (RAO), left anterior oblique (LAO), long-
all diastolic pressures. The atrial pressures increase with inspi- axial oblique, hepatoclavicular (four-chamber), and the caudal
ration rather than decreasing and there is a marked fall in views. All congenital heart disease angiography is performed
[21/6/010/20:42:23] [229–241]
Table 19.5 Commonly Used Angiographic Projections and Cardiac Lesions Best Imaged with These Projections
View Frontal plane angle Lateral plane angle Cardiac lesion best imaged using these planes
Posteroanterior 08 908 Posteroanterior projection for complex heart disease,
and lateral right ventricle in most conditions, pulmonary veins,
lateral for coarctation, pulmonary stenosis and PDA
RAO and LAO 308 RAO 608 LAO RAO useful for mitral valve, RPA, PDA and anterior
VSDs, RAO with 308 cranial angulation useful to see
RPA after a hemi-Fontan; LAO for coarctation, aortic
valve abnormalities
Long-axial oblique 308 RAO 608 LAO and Membranous, outlet, midmuscular, and some apical
208 cranial VSDs, left ventricular outflow tract obstruction, and the
left pulmonary artery
Hepatoclavicular 408 LAO and 1208 LAO and Atrioventricular septal defects, apical and midmuscular
408 cranial 158 cranial VSDs, truncus; with a little less LAO on frontal
projection good to see atrial septal defect
Cranial caudal 208 RAO and 908 and 20–308 Frontal plane: RPA and main pulmonary artery in TOF,
408 cranial caudal lateral for Main pulmonary artery bifurcation in TOF and
after Hemi-Fontan or Bicaval shunt
Laid back 08 and 458 908 Coronary arteries in D-transposition of the great arteries
caudal or double-outlet right ventricle
Abbreviations: RPA, right pulmonary artery; TOF, tetralogy of Fallot; VSD, ventricular septal defect; PDA, patent ductus arteriosis; RAO, right
anterior oblique; LAO, left anterior oblique.
using bi-plane angiography to reduce the amount of contrast

agent given to the individual. Table 19.5 lists the commonly
used angiographic projections.
In addition to selecting the appropriate angiographic
view it is important to remember that the anatomy of any
chamber is best delineated when the chamber is selectively
filled with the appropriate amount of contrast at the appropri-
ate rate. For the best anatomic definition, contrast should be
injected in one second or less. The volume of contrast should be
modified according to both the size and flow rates of the
chamber or vessel being imaged. For example, higher volumes
and flow rates are required to define the ventricular anatomy of
a child with a large intracardiac shunt than a child with
ventricular outflow obstruction. In general, most angiograms
in patients with congenital heart disease use 1.0 to 1.5 cc/kg of
contrast.
Clinical Aspects (Angiography in Specific Lesions)

Ventricular Septal Defects
Membranous defects are best visualized in the long-axial
oblique projection (lateral tube at 60–708 LAO with 308 cranial
angulation). This view elongates the LVOT. This view also is
useful in evaluating the LVOT for possible outflow tract Figure 19.3 Left ventricular angiogram in a 12-month-old child with
multiple VSDs. The angiogram was preformed in the long-axial
obstruction produced by a subaortic membrane and/or possi-
oblique projection left anterior oblique 658 with 308 cranial angulation.
ble prolapse of the aortic valve into the VSD. The companion
In this projection one can demonstrate at least three VSDs. The lower
RAO projection is useful in identifying potential RV outflow defect is in the apical portion of the ventricular septum, the middle
tract obstruction and for making the diagnosis of associated defect is in the midtrabecular portion of the septum and the upper
LV-RA shunts. defect is in the outlet portion of the septum. The long-axial oblique
Muscular VSDs require multiple views depending on the projection profiles well these portions of the atrial septum. Abbrevia-
exact location of the defect or defects. The RAO projection is tion: VSDs, ventricular septal defects.
helpful in profiling the infundibular septum and defects in the
anterior conal septum (subpulmonary defects). The long-axial
oblique projection is useful for malalignment defects and
defects in the midtrabecular portion and some defects in the Atrioventricular Septal Defects
apical portion of the ventricular septum (Fig. 19.3). The hep- The RAO and companion long-axial oblique views are useful in
atoclavicular (four-chamber view) projection is helpful to eval- assessing the degree of atrioventricular valve regurgitation and
uated defects in the posterior trabecular septum, apical septum the status of the LVOT. The hepatoclavicular view is excellent
and around the atrioventricular valves. for evaluation of the inlet (posterior interventricular) septum.
[21/6/010/20:42:23] [229–241]
This view is also well suited to evaluate abnormal position of mitral valve and anterior ventricular septum. The origins of the
the atrioventricular valve attachments and the size of the coronary arteries are best evaluated by performing an aorto-
ventricles. It is also useful in assess the type and severity of gram with marked caudal angulation and with a balloon
atrioventricular valve regurgitation. The companion cranially angiographic catheter positioned antegrade across the aortic
angled RAO projection is useful to evaluate the infundibular valve and with the balloon inflated (Fig. 19.5) (32).
septum and the possibility of RV outflow tract obstruction.
Complex Congenital Defects
The standard posteroanterior and lateral projections are the
Tetralogy of Fallot
initial views for evaluating a child with complex congenital
The RV infundibular narrowing is best visualized by perform-
heart disease (Fig. 19.6). Other views can then be performed to
ing a RV angiogram in a slightly RAO projection with 308 of
define any structures that need better anatomic evaluation. The
cranial angulation with the companion projection being a
echocardiogram can be used to help determine the best angio-
straight lateral (Fig. 19.4). These views provide excellent visu-
graphic view. An angiographic view that is perpendicular to
alization of the anterior deviation of the infundibular septum.
the transduce position and provides the best echocardiographic
The pulmonary arteries are best imaged with the frontal plane
definition of the anatomy is usually a good starting point.
in an RAO projection with cranial angulation and with the
lateral plane in a LAO projection with 308 to 408 of cranial
angulation. The RAO projection defines the right pulmonary Peripheral Pulmonary Artery Anatomy
artery (RPA) and the LAO projection delineates the left pul- In patients with pulmonary atresia the first angiogram is usu-
monary artery. The coronary arteries should always be visual- ally an aortic angiogram in the posteroanterior and lateral
ized in any preoperative patient with tetralogy of Fallot who is projections (Fig. 19.7). Once the source of pulmonary blood
undergoing a heart catheterization. An aortogram positioned in flow is identified, selective angiograms in the aortopulmonary
the RAO and long-axial oblique positions usually result in collaterals can be performed.
adequate delineation of the coronary anatomy. In the neonate, balloon occlusion angiography is useful
(Fig. 19.8) (33). To perform balloon occlusion aortogram, the
aorta is catheterized antegrade with a balloon-tipped angio-
D-Transposition of the Great Arteries catheter. The balloon is inflated immediately prior to contrast
A left ventricular angiogram in the long-axial oblique projec- injection and is then rapidly deflated.
tion is useful to assess the size of the LV, status of the Finally, if pulmonary arteries or segments of the pulmo-
ventricular septum, and the LVOT anatomy. The RAO projec- nary arteries are not visualized on the aortic angiogram, selec-
tion of the left ventricular angiogram is helpful to evaluate the tive pulmonary venous wedge angiograms can be performed
Figure 19.4 RV angiogram from a four-month-old with tetralogy of Fallot who has had a previous right BT shunt. (A) The frontal camera
projection in the RAO cranial projection. The RV infundibular narrowing is marked by the two white arrows, the main, right and left pulmonary
arteries are visualized. One can note stenosis and hypoplasia of the proximal RPA at the site of the BT shunt. The fact that the aorta (AO) is
opacified on the RV injections suggests the presence of a VSD with right-to-left shunting. (B) The companion LAO cranial projection in the
same patient. This projection again demonstrates the infundibular narrowing (solid white arrowhead) and the outlet malalignment VSD (two
smaller white arrows). One can also not that the aorta overrides this VSD. Abbreviations: RV, right ventricle; BT, Blalock Taussig; RAO, right
anterior oblique; LAO, left anterior oblique; LPA, left pulmonary artery; RPA, right pulmonary artery; PA, pulmonary artery; LV, left ventricle.
[21/6/010/20:42:23] [229–241]
Figure 19.5 A laid-back aortogram in a newborn with DTGA. The

frontal camera was positioned with 108 right anterior oblique and 408 Figure 19.7 Aortogram in an infant with pulmonary atresia with
caudal angulation. The LAD arises from the left facing sinus and the VSD. The angiogram was filmed in the posteroanterior projection.
RCA and RCirc arise from the right facing sinus. This is a common The PA are filled by a PDA. Abbreviations: PA, pulmonary arteries;
coronary anomaly observed in transposition. Only the AsAo is AsAo, ascending aorta; PDA, patent ductus arteriosis.
visualized because the balloon of the antegrade angiographic cath-
eter is inflated. One can also note that this child has a right aortic arch
since the first branch is a left innominate artery. Abbreviations: RCA,
right coronary artery; AsAo, ascending aorta; Rcirc, right circumflex
coronary artery; LAD, left anterior descending coronary artery; DTGA,
d-transposition of the great arteries.
Figure 19.8 Balloon occlusion descending aortogram for docu-

mentation of the source of pulmonary blood flow in a neonate with
pulmonary atresia with ventricular septal defect. The balloon-tipped
Figure 19.6 Left ventricular angiogram of a child with tricuspid angiographic catheters has been advanced from the right ventricle
atresia. The angiogram was performed in the posteroanterior pro- into the ascending and then the DsAo. Contrast injected behind the
jection. The LV opacifies the hypoplastic RV infundibular chamber inflated balloon (arrow) produces dense opacification of single large
(a line outlines the chamber) and PA. There is no apical portion of collateral arising from the mid thoracic aorta that opacifies confluent
the right ventricle identified. Abbreviations: LV, left ventricle; RV, right and left pulmonary arteries. One should also note that the
right ventricle; PA, pulmonary artery. infant has a right aortic arch. Abbreviation: DsAo, descending aorta.
[21/6/010/20:42:23] [229–241]
Figure 19.9 Series of angiograms performed to identify the pulmonary artery anatomy in a neonate with pulmonary atresia with ventricular
septal defect. (A) An aortogram that fails to identify the pulmonary arteries. (B) A wedge pulmonary vein injection. The catheter is wedged in
the LPV, contrast injected refluxes into the LPA. Although the LPA is well visualized there is only a hint of a confluence and the RPA is not
visualized. (C) A second wedge pulmonary vein injection. The catheter is wedged in the RUPV, contrast injected refluxes into RPA, outlining a
confluence as well as the LPA. Abbreviations: LPV, left middle pulmonary vein; LPA, left pulmonary artery; RUPV, right upper pulmonary vein;
RPA, right pulmonary artery.
(Fig. 19.9). The technique involves advancing an end-hole and the contrast/saline is injected by hand as fast as possible.
catheter across the atrial septum into the pulmonary vein that Serious complications from this technique have been reported
is best for delineating the portion of the pulmonary tree that with the use of a nonballoon end-hole catheter (34).
could not previously be visualized. The catheter should then be
advanced until it is wedged in the pulmonary vein. If the
catheter has a balloon tip, the balloon can be inflated and CONCLUSION
wedge position confirmed by the injection of a small amount Although advances in noninvasive imaging have reduced the
of contrast. A 10-cc syringe in which contrast and saline have role of diagnostic cardiac catheterization in the patient with
been layered such that the saline is drawn up first and contrast congenital heart disease, recent developments in transcatheter
second is used for the injection. The catheter is then wedged interventions as well as marked advancements in surgical
[21/6/010/20:42:23] [229–241]
management of many forms of extremely complex congenital 15. Van Slyke DD, Neill JM. Blood gases I. J Biol Chem 1924; 61:524–584.
heart disease, have resulted in diagnostic cardiac catheteriza- 16. Fick A. Uber die messung des blutquantums in den herzventri-
tion remaining a critical component of the evaluation and keln. Sits der Physik-Med ges Wurtzberg 1870:16.
treatment of children with congenital heart disease. This chap- 17. Freed MD, Keane JF. Cardiac output measured by thermodilution
in infants and children. J Pediatr 1978; 92:39–42.
ter has summarized four of the areas that are critical to know
18. Lister G, Hoffman JI, Rudolph AM. Oxygen uptake in infants and
when performing a cardiac catheterization in a pediatric children: a simple method for measurement. Pediatrics 1974; 53:
patient: sedation for the procedure, vascular access, hemody- 656–662.
namic assessment, and angiography. 19. LaFarge CG, Miettinen OS. The estimation of oxygen consump-
tion. Cardiovasc Res 1970; 4:23–30.
20. Barratt-Boyes BG, Wood EH. The oxygen saturation of blood in
REFERENCES the venae cavae, right-heart chambers, and pulmonary vessels of
1. Cote CJ, Wilson S. Guidelines for monitoring and management of healthy subjects. J Lab Clin Med 1957; 50:93–106.
pediatric patients during and after sedation for diagnostic and 21. Freed MD, Miettinen OS, Nadas AS. Oximetric detection of intra-
therapeutic procedures: an update. Pediatrics 2006; 118:2587–2602. cardiac left-to-right shunts. Br Heart J 1979; 42:690–694.
2. American Academy of Pediatrics Committee on Drugs and Com- 22. Dexter L, Haynes FW, Burwell CS, et al. Studies of congenital
mittee on Environmental Health: use of chloral hydrate for seda- heart disease. II. The pressure and oxygen content of blood in the
tion in children. Pediatrics 1993; 92:471–473. right auricle, right ventricle, and pulmonary artery in control
3. Meretoja OA, Rautiainen P. Alfentanil and fentanyl sedation in patients, with observations on the oxygen saturation and source
infants and small children during cardiac catheterization. Can J of pulmonary “capillary” blood. J Clin Invest 1947; 26:554–560.
Anaesth 1990; 37:624–628. 23. Cannon BC, Feltes TF, Fraley JK, et al. Nitric oxide in the evalu-
4. Berman W Jr., Fripp RR, Rubler M, et al. Hemodynamic effects of ation of congenital heart disease with pulmonary hypertension:
ketamine in children undergoing cardiac catheterization. Pediatr factors related to nitric oxide response. Pediatr Cardiol 2005; 26:
Cardiol 1990; 11:72–76. 565–569.
5. Tosun Z, Akin A, Guler G, et al. Dexmedetomidine-ketamine and 24. Scheurer M, Bandisode V, Atz AM. Simplified pulmonary vaso-
propofol-ketamine combinations for anesthesia in spontaneously dilatory testing in the cardiac catheterization laboratory with nasal
breathing pediatric patients undergoing cardiac catheterization. J cannula nitric oxide. Pediatr Cardiol 2006; 27:84–86.
Cardiothorac Vasc Anesth 2006; 20:515–519. 25. Atz AM, Adatia I, Lock JE, et al. Combined effects of nitric oxide
6. Seldinger SI. Catheter replacement of the needle in percutaneous and oxygen during acute pulmonary vasodilator testing. J Am
arteriography; a new technique. Acta radiol 1953; 39:368–376. Coll Cardiol 1999; 33:813–819.
7. Takahashi M, Petry EL, Lurie PR, et al. Percutaneous heart catheter- 26. Rosenzweig EB, Widlitz AC, Barst RJ. Pulmonary arterial hyper-
ization in infants and children. I. Catheter placement and manip- tension in children. Pediatr Pulmonol 2004; 38:2–22.
ulation with guide wires. Circulation 1970; 42:1037–1048. 27. Gorlin R, Gorlin SG. Hydraulic formula for calculation of the area
8. Carlson KM, Rutledge JM, Parker BR, et al. Use of tissue plasmi- of the stenotic mitral valve, other cardiac valves, and central
nogen activator for femoral artery thrombosis following trans- circulatory shunts. I. Am Heart J 1951; 41:1–29.
catheter coil occlusion of patent ductus arteriosus. Pediatr Cardiol 28. Bache RJ, Wang Y, Jorgensen CR. Hemodynamic effects of exercise
2005; 26:83–86. in isolated valvular aortic stenosis. Circulation 1971; 44:1003–1013.
9. Ino T, Benson LN, Freedom RM, et al. Thrombolytic therapy for 29. Chun PK, Rocchini AP. Occult constrictive pericarditis in infancy.
femoral artery thrombosis after pediatric cardiac catheterization. Documentation by rapid volume expansion. Chest 1980; 78:648–650.
Am Heart J 1988; 115:633–639. 30. Bargeron LM Jr., Elliott LP, Soto B, et al. Axial cineangiography in
10. Vitiello R, McCrindle BW, Nykanen D, et al. Complications congenital heart disease. Section I. Concept, technical and ana-
associated with pediatric cardiac catheterization. J Am Coll Car- tomic considerations. Circulation 1977; 56:1075–1083.
diol 1998; 32:1433–1440. 31. Elliott LP, Bargeron LM Jr., Bream PR, et al. Axial cineangiogra-
11. Luedde M, Krumsdorf U, Zehelein J, et al. Treatment of iatrogenic phy in congenital heart disease. Section II. Specific lesions. Circu-
femoral pseudoaneurysm by ultrasound-guided compression lation 1977; 56:1048–1093.
therapy and thrombin injection. Angiology 2007; 58:435–439. 32. Mandell VS, Lock JE, Mayer JE, et al. The “laid-back” aortogram:
12. Shim D, Lloyd TR, Beekman RH III. Transhepatic therapeutic an improved angiographic view for demonstration of coronary
cardiac catheterization: a new option for the pediatric interven- arteries in transposition of the great arteries. Am J Cardiol 1990; 65:
tionalist. Catheter Cardiovasc Interv 1999; 47:41–45. 1379–1383.
13. Lim DS, Ragosta M, Dent JM. Percutaneous transthoracic ventric- 33. Keane JF, McFaul R, Fellows K, et al. Balloon occlusion angiog-
ular puncture for diagnostic and interventional catheterization. raphy in infancy: methods, uses and limitations. Am J Cardiol
Catheter Cardiovasc Interv 2008; 71:915–918. 1985; 56:495–497.
14. Maher KO, Murdison KA, Norwood WI Jr., et al. Transthoracic 34. Alpert BS, Culham JA. A severe complication of pulmonary vein
access for cardiac catheterization. Catheter Cardiovasc Interv 2004; angiography. Br Heart J 1979; 41:727–729.
63:72–77.
[24/6/010/13:3:7] [242–252]
20
Cardiac catheterization for the adult with complex

congenital heart disease
John F. Rhodes, Jr. and Jorge R. Alegrı́a B.
INTRODUCTION FUNDAMENTALS FOR CONGENITAL

The incidence of congenital heart disease is 0.8% (8/1000) live CARDIAC CATH
births (1). These patients have a cardiovascular abnormality Hemodynamic Assessment: Oxygen Saturations
that is secondary to an altered embryonic development with Obtained in Room Air
abnormal shunting patterns that result in further problems in Oxymetry can detect shunts and be utilized in the calculation of
the structure and function of the cardiovascular system. cardiac output. It is critical to avoid obtaining the data with the
Currently, there are more adults than children in the United patient on nasal canulae oxygen. It is important to obtain the
States with congenital heart disease making this chapter of saturation date while the patient is in room air and the satura-
critical importance to the cardiologist (2). Many of these tion should be drawn from a location distal to the shunt lesion.
patients with adult congenital heart disease (ACHD) will Oxygen saturation is the percentage of hemoglobin that is pres-
require additional catheter-based procedures that are either ent as oxyhemoglobin and is measured by reflectance. Oxygen
diagnostic to assess the complex anatomy or interventional content is the total amount of oxygen present in the blood. This
procedures that allow therapeutic options to the patient. Over includes the oxyhemoglobin plus the oxygen dissolved in the
the past few decades, with the evolution of noninvasive plasma. The oxygen content is calculated with the following
imaging techniques including real-time 3-D echocardiography formula:
as well as magnetic resonance and CT angiography, the
O2 content ¼ ðO2 saturation 1:36 10 Hgb concentrationÞ
catheterization laboratory is now used more often for inter-
ventional procedures. The objective of this chapter is to pro- In this formula 1.36 represents the amount of O2 1 g of
vide the cardiologist with a general overview of complex hemoglobin carries when fully saturated. A left-to-right shunt
ACHD and its management options in the congenital cardiac is diagnosed when a significant oxygen step-up is found. Spe-
catheterization laboratory. cifically, a step-up of >11% from the SVC to RA indicates a left-
to-right shunt at the atrial level, >7% from the RA to the right
ventricle (RV) is consistent with a shunt at the RV level, >5%
ANATOMIC AND PRECATH CONSIDERATIONS from the RV to the pulmonary artery indicates a systemic to
Congenital cardiac catheterization is performed with either pulmonary artery shunt, and 9% from the SVC to the pulmo-
general anesthesia or moderate sedation with a combination of nary artery has a high predictive accuracy to detect a 2:1 left-to-
opiates and benzodiazepines. Percutaneous access using stan- right shunt.
dard Seldinger technique in the femoral artery and/or vein is To calculate blood flows, oxygen contents and oxygen
performed and alternative venous access could include the consumption are used.
internal jugular, subclavian or transhepatic approaches. Arte- VO2 ðmL=minÞ
rial access may also be obtained from the femoral and radial Systemic blood flowðQs Þ ¼
SA O2 content MV O2 content
arteries. Hemodynamic measurements are obtained initially
always in room air precluding the need for a pulmonary vein VO2 ðmL=minÞ
PO2. Subsequently, angiograms will further delineate the Pulmonary blood flowðQp Þ ¼
PV O2 content PA O2 content
anatomy or lesion severity. For therapeutic procedures, a
catheter is passed across the target area, such as stenosis or where SA is systemic arterial saturation, MV mixed venous, PV
abnormal shunt. A guide wire is then passed through the pulmonary vein, and PA pulmonary artery saturation.
catheter to provide a track over which therapeutic devices are The mixed venous saturation is calculated with the fol-
delivered. Balloon catheters are threaded directly, whereas lowing formula:
stents and occlusion devices are protected or constrained
ð2SVC þ 1IVCÞ
within a long delivery sheath. Anticoagulation is managed Mixed venous saturation ðMVÞ ¼
3
with heparin 100 units per kg body weight (maximum
5000–7000 units) for a goal activated clotting time (ACT) Another flow that is calculated in the study of congenital
200 to 250 seconds, depending on the procedure to be per- heart defects in the cardiac catheterization laboratory is the
formed. In addition, antibiotics are given to patients who effective flow (Qe), that is, the quantity of systemically mixed
receive an implanted device. venous blood that circulates and is oxygenated in the lungs and
[24/6/010/13:3:7] [242–252]
CARDIAC CATHETERIZATION FOR THE ADULTS WITH CONGENITAL HEART DISEASES 243
then circulates through the systemic capillaries. On the basis of

this, the effective flow is calculated using the pulmonary vein
and mixed venous saturations, and in the absence of shunts
equals Qp = Qs = Qe.
O2 consumption
Effective pulmonary flow ¼
PV O2 content MV O2 content
In the absence of shunting, the mixed venous saturation
is equal to the pulmonary artery saturation. The left-to-right
shunt (flow) is obtained by subtracting the effective pulmonary
blood flow from the total pulmonary blood flow: left-to-right
shunt = Qp Qe.
Hemodynamic Assessment: Intracardiac

and Intravascular Shunts
Shunts allow the intermixing of saturated blood with unsatu-
rated blood. Various shunts can be either cardiac or vascular
and are classified as left to right, right to left, and bidirectional.
Left-to-right shunting results in increased pulmonary blood
flow. Right-to-left shunt results in a lower arterial saturation,
generally less than 95%. In bidirectional shunts the left-to-right
shunting is calculated as Qp Qe and the right-to-left shunting
as Qs Qe. In patients with residual shunts and pacemaker
leads there is a >2-fold increase in thromboembolic events (3).
Hemodynamic Assessment: Pressure

Data (Room Air)
Normally in the right atrial pressure tracing the “a” wave is
dominant and in the left atrial pressure tracing the “v” wave is Figure 20.1 Normal anatomy and hemodynamics. The saturations
show no step-up to suggest a shunt and in parentheses the percent
dominant. In secundum atrial septal defect there is a larger v
step that would be necessary for a shunt to be considered beyond
wave in the right atrial pressure (equal a and v waves). Eleva- measurement error.
tion of the a wave can be seen in pulmonary stenosis. Large
“cV” waves are seen in tricuspid insufficiency. Rise in right
ventricular pressure is seen in outflow tract obstruction or
pulmonary hypertension. Ventricular septal defects occasion- stenoses. When doing these angiograms the levophase angio-
ally are close and are associated with anomalous and hyper- gram is useful to identify the pulmonary vein drainage and
trophied muscle bands, creating the so-called double-chamber additional left-sided structures and function. Pulmonary wedge
RV, which results in a proximal chamber with elevated pres- angiogram is occasionally used to assess the pulmonary vascular
sure and a distal one with low pressure in the RV. Abnormal anatomy and its changes when there is important pulmonary
aortic tracings can be the result of abnormal gradients such as vascular obstructive disease. In congenital heart diseases
in coarctation of the aorta or supravalvular stenosis. A wide biplane imaging should be used. To visualize the atrial septum
pulse pressure is seen in aortic insufficiency, aortopulmonary a cranial 458, left anterior oblique (LAO) 458 view is useful. To
shunts or systemic AVM. see the transverse arch and the area of aortic coarctation a direct
lateral view if a biplane laboratory and a steep caudal/LAO
view are performed (Fig. 20.2).
Hemodynamic Assessment: Angiography
To understand congenital heart diseases it is necessary to be
familiar with normal features of the different cardiac segments CATHETERIZATION FOR COMPLEX
(Fig. 20.1). Angiography is performed and each structure is CONGENITAL HEART DEFECTS
evaluated for anatomy, function and connections. Contrast Patent Ductus Arteriosus
injection of the innominate vein is done to rule a persistent Patent ductus arteriosus (PDA) is a cardiac derangement usu-
left superior vena cava or other anomalies. Inferior vena cava ally detected early in life and occasionally during adulthood.
venograms are needed in patients with Fontan circulation to The finding of an audible PDA is generally regarded as an
assess for inferior baffle obstruction and shunts resulting in indication for treatment since patients with an audible PDA
systemic hypoxemia. Right ventriculogram is frequently per- typically have a shortened life expectancy (4) and a potential
formed for either function or anatomy. If for function the increase risk for endarteritis. Survival with an audible PDA to
injection rate can be slowed down to see the systolic function an advanced age is unusual without significant morbidity (5).
over several heart beats but for anatomy the injection should be However, with a small or restrictive PDA clinical findings such
approximately 0.5 to 1 cc/kg over a single heart beat. This is as murmur and clinical symptoms are uncommon other than a
usually 30 to 40 cc contrast at 30 to 40 cc/sec when the heart potentially higher risk of endarteritis compared with the gen-
rate is 60 beats/min. Pulmonary angiography is used to eval- eral population. Historically, over half of the mortalities from
uate pulmonary valve regurgitation or branch pulmonary PDA in the pre-antibiotic era were from endarteritis rather than
[24/6/010/13:3:7] [242–252]
Figure 20.2 (A) The transverse arch and isthmus are best imaged by positioning the lateral camera in a direct lateral view and (B) the AP
camera in a steep caudal and left anterior oblique view.
heart failure (6). In the presence of significant shunting, con-

gestive heart failure initially and pulmonary hypertension later
can occur. In addition, by 40 years of age nearly a third of
patients with an unrepaired PDA develop heart failure, pul-
monary hypertension, or endarteritis and two thirds do not
survive beyond the fifth decade (7). Therefore, the current
standard of care is surgical or device PDA closure when a
PDA is auscultated (8). In the ACHD patient, coexistent path-
ologies like dilation of the aorta secondary to the chronic left-to-
right shunt, calcified atheromatous lesions, and aneurysm of
the ductus may result in surgical correction requiring cardio-
pulmonary bypass (9). Consequently, percutaneous occlusion
of the PDA as an excellent lower risk alternative for the ACHD
patient and precludes the need for a potentially high-risk
cardiopulmonary bypass procedure (Fig. 20.3) (10).
Both a femoral venous and femoral arterial sheath is
placed to perform percutaneous PDA closure and patients are
given local anesthesia, moderate sedation, and heparin. If pos-
sible the PDA is crossed retrograde from the pulmonary artery
into the aorta for the purpose of coil or device delivery. When
the ductus can only be crossed antegrade (from the aorta),
either the coil may be deployed in that fashion or the delivery Figure 20.3 Lateral projection of a patent ductus arteriosus.
catheter may be advanced retrograde by using a snare and rail
technique (11). A biplane aortogram in the 40 RAO (AP camera)
and direct lateral is preferred but if the lab is single plane the
lateral projection or steep LAO is utilized. This angiogram is
used to delineate the location, anatomy, and narrowest diam- residual shunts. The goal is to achieve complete closure of the
eter of the PDA. A single or multiple 0.038-in Gianturco coils ductus by angiography prior to leaving the catheterization
are deployed in the PDA using a tapered catheter for enhanced laboratory. If residual shunt, other than slow filtration through
control as previously described (12). The goal is to deploy 1 the device, is seen on angiography, the PDA is crossed again
loop of the coil in the main pulmonary artery and the remain- and additional coils can be deployed in the same fashion.
ing loops in the ductal ampulla. More recently, many centers Prophylactic antibiotics are administered at the time of coil
have moved toward using the Amplatzer duct occluder (AGA delivery and for 24 hours thereafter. The risk for adverse events
Medical Corporation, Golden Valley, Minnesota, U.S.) almost is low with reported complications including coil or device
exclusively for PDA closure in adults (13). A repeat aortogram embolization, hemolysis from high-velocity residual shunting
is performed to evaluate for coil or device position and for and infection. Adult patients can present with a calcified PDA
[24/6/010/13:3:7] [242–252]
Figure 20.4 (A) Cineangiogram in the direct AP projection demonstrating a persistent left superior vena cava to the left atrium with a
persistent left innominate vein and (B) the same projections after percutaneous closure from the left internal jugular vein using an Amplatzer
vascular plug II.
and are at higher risk for surgical closure when calcification is

present.
Chest X ray (CXR) and transthoracic echocardiography
are performed before discharge the next morning. Standard
endarteritis prophylaxis is recommended for six months after
closure to allow coil or device endothelialization to occur and
echocardiography should confirm complete closure prior to
releasing the patient from routine follow-up with cardiology.
Persistent Left Superior Vena Cava

with Hypoxemia
This is a common congenital anomaly of the venous system. It
is relevant to be aware of this malformation before pacemaker
lead placement from the left subclavian vein approach. The
most common anatomy is absence or hypoplasia of the left
innominate vein and a direct connection of the left internal
jugular vein and left subclavian vein to either the left atrial
appendage to the left atrium or through the coronary sinus to
the right atrium. The coronary sinus anatomy should be sus-
pected when a prominent coronary sinus ostia is seen by
echocardiography or MRI. The direct connection to the left Figure 20.5 The inferior vena cava below the hepatic veins is
atrium (Fig. 20.4) should be suspected when patients present interrupted with subsequent systemic venous drainage via the
with resulting in systemic hypoxemia or a history of brain azygous vein to the left-sided superior vena cava.
abscess after dental procedures.
Interrupted Inferior Vena Cava

This anomaly is commonly but not always associated with
complex congenital heart diseases or forms of heterotaxy. The Myocardial Bridging
inferior vena cava below the hepatic veins but above the renal The intramural course of a coronary is very common congenital
veins is absent and thus the venous return is interrupted with coronary anomaly although its clinical significance remains to
subsequent systemic venous drainage via the azygous vein to be elucidated. In selected cases percutaneous or surgical inter-
the superior vena cava (Fig. 20.5). The hepatic veins enter the vention is warranted (generally severe bridging and evidence
right atrium directly but although usually together they can of ischemia) but in the majority with adequate antegrade flow
enter contra lateral to each other. during diastole do not warrant any intervention (Fig. 20.6).
[24/6/010/13:3:7] [242–252]
Figure 20.6 Myocardial bridging of the left circumflex coronary artery (A) during systole and (B) during diastole.
Anomalous Coronary Arteries

They can occur isolated or in conjunction with other structural
heart disease abnormalities. The recognition and evaluation of
coronary anomalies has become a very important part of the
evaluation of congenital heart diseases. The left coronary artery
arises from the left sinus of Valsalva the right coronary artery
from the right sinus of Valsalva. The right or left coronary
artery may arise from inappropriate sinus of Valsalva, the most
common one is when the left circumflex artery arises from the
right coronary sinus. Other types includes anomalous right
coronary artery from the left coronary sinus, single coronary
artery, anomalous left coronary artery from the pulmonary
artery (ALCAPA) also called Bland-White-Garland syndrome.
ALCAPA is characterized by pathologic q waves in leads I,
aVL, and precordial leads V4 to V6. It is possible to find a small
left-to-right shunt at the pulmonary level. The left ventriculo-
gram demonstrates left ventricular dysfunction (anterolateral
wall) and mitral valve regurgitation. Aortogram will demon-
strates a large right coronary artery and filling of the left
coronary by collaterals from the RCA with passage of contrast
from the left main to the pulmonary artery.
Figure 20.7 Lateral projection of a coronary fistula from the prox-
imal right coronary artery to the superior vena cava resulting after
Coronary Fistulae transcatheter intervention with an Amplatzer vascular plug II.
They commonly arise from the proximal portion of the native
coronary artery and enter either the pulmonary artery or the
atria. Coronary fistulae can produce ischemia due to steal of
coronary perfusion. If large enough they can produce a left-to-
coarctation may represent one aspect of more diffuse arterio-
right shunt. Transcatheter intervention is indicated when there
pathy. Late complications can include aneurysm formation and
is coronary steal and symptoms (Fig. 20.7).
dissection of the ascending aorta rather than the region of prior
repair or intervention. Cerebrovascular events from rupture of
Coarctation of the Aorta berry aneurysms may occur before or after surgical repair of
Coarctation of the aorta has been estimated to occur with a coarctation. Persistent hypertension has been reported despite
frequency of 7% to 9% (14). It is usually diagnosed in infancy or apparently complete relief of obstruction either surgically or by
early childhood but can go undetected to become apparent well angioplasty. Life expectancy beyond the sixth decade is
into adulthood. The natural history suggests that isolated unusual if the coarctation is not repaired with a mean survival
[24/6/010/13:3:7] [242–252]
Current surgical practice warrants early repair, usually within

the first year of life. Without surgical intervention, only about
10% of patients survive beyond the age of 20 years. Adults with
repaired TOF usually have undergone at least one surgical
procedure but sometimes two to three procedures before their
twenties. The repair involves patch closure of the VSD, variable
degrees of RV outflow tract resection, and reconstruction, pul-
monary valvotomy, or placement of an RV to pulmonary artery
conduit, either bioprosthetic or homograft. Distal branch pul-
monary arterial stenosis may have been repaired, or residual
lesions may be present. Examination focuses on the detection of
residual lesions. Cardiac catheterization can be used for diag-
nostic and potentially for interventional purposes. Residual
shunts are actively sought at the atrial, ventricular, and pul-
monary arterial levels. RV pressure is systemic in a patient who
has not undergone surgical repair. After surgical repair, ele-
vated RV pressure supports the presence of residual obstruc-
tive lesions, the levels of which are documented. Careful
pullback recordings are performed from the branch pulmonary
arteries to the RV because stenosis at each level is possible. The
presence of stenosis at a prior shunt site is expected. RV end
Figure 20.8 Stent angioplasty for native coarctation of the aorta. diastolic pressures may be elevated in the setting of pulmonary
insufficiency. Left heart catheterization is performed if non-
invasive studies suggest residual VSDs. Angiography includes
a cranialized right ventriculography and possibly selective
of approximately 35 years (15). The coarctation site is typically pulmonary arterial injections if hemodynamic findings suggest
just beyond the origin of the left subclavian artery across from stenosis. Left ventriculography better demonstrates residual
the ampulla of the ductus arteriosus. Collateral circulation VSDs in the presence of sub systemic RV pressures. Aortic
often is present in older patients to bypass the coarctation root injection demonstrates the presence of aortic insufficiency,
and provide blood flow to the lower extremities. The most confirms the presence of grossly abnormal coronary artery
common origins of these vessels are from the subclavian origins or branching patterns, and reveals prior surgical shunts
arteries through the internal thoracic arteries and the thyrocer- or aortopulmonary collateral vessels. If present, shunts and
vical and costocervical branches. These vessels communicate collateral vessels are best visualized in the posteroanterior and
with the intercostal arteries, which then perfuse the descending lateral projections after selective injection by hand. Selective
aorta distal to the coarctation. This can produce diminished but coronary arteriography is recommended in the care of adult
palpable lower extremity pulses and mask substantial coarcta- patients to exclude coronary artery disease before surgical re-
tion. MRI is particularly helpful in the evaluation of coarctation intervention. In a small proportion of patients the left anterior
among adult patients and cine Careful hemodynamic assess- descending coronary artery can arise from the right coronary
ment across the coarctation is essential to find a peak systolic cusp. This is important because a valved-RV to pulmonary
gradient. Intervention typically is indicated when the peak artery conduit should be used since right ventriculotomy
gradient is greater than 20 to 30 mmHg, but other clinical entails the risk of injuring the LAD. Long-term follow-up
factors must also be considered (congestive heart failure, left studies have demonstrated that patients with TOF will have
ventricular hypertrophy, abnormal stress test result, uncon- right ventricular enlargement and dysfunction due to chronic
trolled upper extremity hypertension). Also, the gradient may volume overload secondary to severe pulmonary valve regur-
be artificially decreased in the presence of collateral vessels. gitation. Pulmonary valve replacement is often indicated. An
In general coarctation can be divided into recurrent or evolving interest in performing percutaneous pulmonary valve
native obstruction. Although recurrent coarctation is now being replacement exists, although further data is needed.
addressed with angioplasty or stent angioplasty routinely,
native coarctation should still be approached cautiously. For
selected older patients, stent placement may be considered as Right Ventricular Outflow Tract Obstruction: Native
the primary intervention for native coarctation but should be or Postoperative
considered in light of the need for growth of the aorta and risk When severe a large a wave will be seen in the right atrial tracing
for aneurysm formation. However, the long-term outcome into because of decreased compliance of the RV. A gradient will be
late adulthood after stent placement has yet to be determined, found in the right ventricular outflow tract. Right ventricular
and investigation continues (Fig. 20.8). angiography should be performed with AP and lateral views.
The valve will be thickened and with restricted motion (dom-
Tetralogy of Fallot ing). Usually infundibular hypertrophy is present. On the basis
Tetralogy of Fallot (TOF) occurs in approximately 6% of new- of the peak gradient it can be classified as mild (<30 mmHg),
borns found to have congenital heart disease and it is the most moderate (30–49 mmHg) and severe (>50 mmHg). Pulmonary
common form of cyanotic heart disease among adult patients valve can be dysplastic. Dysplastic pulmonary valve is present in
with congenital heart disease (16). It consists of a large ven- the Noonan Syndrome.
tricular septal defect, overriding aorta, pulmonary stenosis For non-dysplastic valves the treatment of choice is
(infundibular or valvular), and right ventricular hypertrophy. percutaneous pulmonary balloon valvuloplasty. Because of
[24/6/010/13:3:7] [242–252]
Figure 20.9 Percutaneous pulmonary stent valve placement for severe insufficiency of a surgically placed homograft (A) prior to intervention
and (B) after placement of the Edwards stent valve.
infundibular hypertrophy, immediately post valvuloplasty the artery can result in severe systemic hypoxemia in the presence of
gradient can increase (dynamic obstruction). In some rare cases a patent foramen ovale or atrial level shunt because of sudden
there is severe increase in the gradient immediately post compliance change within the RV.
valvuloplasty, with hypotension and hypoxemia (“suicidal
RV”). Infundibular contraction can be decreased with b-blocker D-Transposition of the Great Arteries
therapy. D-Transposition is a malalignment of the great vessels with the
More recently, ongoing clinical trials are looking at per- pulmonary artery arising from the morphologic left ventricle
cutaneous pulmonary stent valve placement for severe insuffi- (LV) and the aorta arising from the morphologic RV. Today, all
ciency of a surgically placed homograft. The two clinical trials newborns with this form of CHD and many teenagers have been
include the Melody valve (Metronic BV, Heerlen, Limburg, The treated with arterial switch operation, described by Dr. Janten
Netherlands) and the Edwards stent valve (Edwards Life- (18). Unfortunately though, most of the young adults evaluated
science, Irvine, California, U.S.) (Fig. 20.9). Although prelimi- with this form of CHD were treated earlier with an atrial rather
nary data from Europe and via these two FDA approved than arterial switch known as either a Mustard or Senning oper-
feasibility studies, long-term data remains pending (17). ation depending on if the atrial baffle was made of pericardial
tissue or Dacron material (19). The main hemodynamic and thus
cardiac catheterization concerns for these patients as adults
Branch Pulmonary Stenoses or Distortion
include the systemic RV function because of the poor morphol-
For the most part, transcatheter therapies in the care of adult
ogy of the RV for the systemic work load, baffle leaks predom-
patients with TOF are limited to patients who have undergone
inately within the connection of the superior vena cava baffle to
surgical treatment with attention to residual obstructive lesions
the right atrium resulting in a right-to-left shunt, systemic hypo-
in the main pulmonary artery, RV to pulmonary artery conduit,
xemia or systemic embolic events, baffle obstruction resulting in
or distal pulmonary arteries. Prior shunt sites may become
superior vena cava syndrome or enlargement of the azygous vein
stenotic with time and necessitate balloon angioplasty and
as it “pops” off to the inferior vena cava and arrhythmias espe-
possibly stent placement. Success has been achieved in these
cially sick sinus syndrome and the need for a pacemaker inser-
situations, but many residual lesions necessitate surgical re-
tion. Non-interventional catheter-based evaluation is reserved
intervention. Hypoplasia or stenosis of branch pulmonary
mainly for hemodynamic interpretation of the failing systemic
arteries will result in an increased afterload to the RV and
RV to obtain an end diastolic pressure, filling pressures, and
hypoperfusion to one lung or segments, with subsequent over-
pulmonary vascular resistance. These data are critical prior to
circulation to other segments. Branch pulmonary stenosis can
proceeding for a possible cardiac transplantation.
be congenital in origin or acquired as a complication of surgical
The patient with multiple baffles leak is often the most
interventions. In adults they are treated with stenting.
difficult intervention to undertake. These patients have very
irregular and abnormal connections within the baffle and the
Sinus of Valsalva Aneurysms morphologic right atrium and right atrial appendage. Techniques
The sinus of Valsalva aneurysm can potentially rupture and previously described (20) for this situation include Gianturco
produce a left-to-right shunt and right-sided volume overload. coils (Cook, Bloomington, Indiana, U.S.), Amplatzer vascular
In addition, postoperative adverse events with the right coronary plugs (AGA Medical, Minneapolis, Minnesota, U.S.), and
[24/6/010/13:3:7] [242–252]
Amplatzer ASD and PDA devices (AGA Medical, Minneapolis, is as portion of the RV that is “atrialized.” Associated lesions
Minnesota, U.S.). More recently a few centers have used the Gore include PFO, ASD, and pulmonary stenosis (23).
cuffed “covered” stents (WL Gore, Flagstaff, Arkansas, U.S.)
within the SVC baffle to essentially exclude the baffle leak (21).
The patient with SVC baffle obstruction will require stent
Single-Ventricle Physiology: Postoperative
angioplasty with balloon expandable peripheral diameter
Physiology
A patient with a functional single ventricle typically proceeds
stents (12–18 mm). The position of the stent is important to
with palliation, including the systemic shunt (Norwood oper-
avoid too much overlap of the inferior baffle but long enough to
ation, Potts shunt, Waterston shunt, or other central shunts),
include the more distal SVC. Often a second stent more distal
followed by the superior vena caval shunt or Glenn operation,
may be necessary. We recommend considering the balloon-in-
and ultimately the completion of the caval pulmonary shunt
balloon catheter so these positioning issues can be addressed.
with the Fontan operation (Fig. 20.10). Examples of congenital
Also, the lesion often can be either scarred and resistant to the
defects that may necessitate such palliation include hypoplastic
balloon or compliant and more of a kink in the baffle. We
left heart, tricuspid atresia, pulmonary atresia with intact
therefore often use a soft noncompliant balloon to “test” the
ventricular septum, or unbalanced complete AV canal. The
lesion resistance measuring in the AP and lateral projections
adult congenital patient with single-ventricle physiology is
and then choosing an angioplasty-stent and balloon that will be
usually palliated with a Fontan but few patients remain with
large enough to adhere well to the walls and not dislodge
single lung supplied by a superior vena cava to the pulmonary
toward the atrioventricular (AV) valves. If a patient already has
artery shunt (Glenn) or a systemic artery to pulmonary artery
pacing wires across the baffle a temporary wire is placed from
shunt. The Fontan procedure represents the final palliative
the lower extremity and the pacing wires are then removed
procedure for single-ventricle physiologic status. This proce-
allowing the stent to be placed. Once the stent is fully dilated
dure completes the direction of the remaining systemic venous
the pacing wires are replaced across the superior vena cava and
blood from the inferior vena cava and hepatic veins to the
the lower extremity temporary wire is subsequently removed.
pulmonary arteries. This is accomplished in most cases by
The final possible derangement to discuss is baffle
means of either an external conduit or an intra-atrial lateral
obstruction in the pulmonary venous baffle. This obstruction
tunnel, which courses from the lateral and inferior aspect of the
is quite difficult to diagnose and is often within the midpulmo-
right atrium (24). The atrial appendage or superior vena caval
nary venous baffle between the entrance of the pulmonary
stump transected during the Glenn procedure is directed to the
veins to the morphologic left atrium and the tricuspid valve
pulmonary artery, effectively “septating” the circulation. Pul-
apparatus. Multiple imaging modalities including MRI, CT
monary blood flow is achieved passively without the assistance
angiography or cardiac catheterization with pulmonary artery
of a ventricular pumping chamber. For this reason it is imper-
wedge pressures and angiograms are utilized to delineate this
ative to have low pulmonary pressures and vascular resistance.
abnormality. Despite these techniques, it remains difficult to
Important derangements include the Fontan pathway
image thus if the wedge pressures are elevated compared with
obstruction, persistent fenestration, or venovenous fistulae
the RV end diastolic pressures we use a pigtail and floppy wire
both of the later resulting in systemic hypoxemia and increased
to position a hemodynamic catheter retrograde from the aorta
risk of cerebrovascular events.
to the RV across the tricuspid valve and into the morphologic
left atrium. Care is taken to assure the catheter is positioned
near the pulmonary veins and thus across the midpulmonary
baffle obstruction. Once the hemodynamic data is obtained this
catheter is exchanged for an angiographic catheter to perform a
power injection directly in the morphologic left atrium. The
main importance of these maneuvers is that long standing
pulmonary venous obstructions can result in pulmonary hyper-
tension or elevated pulmonary vascular resistance and subse-
quently make the patient possibly a poor candidate for cardiac
transplantation.
Congenitally Corrected Transposition

of the Great Arteries
This defect is characterized by AV discordance and ventricu-
loarterial discordance. The RV is the systemic ventricle and
with time will develop left ventricular dysfunction. The tricus-
pid valve is the systemic AV valve and can develop significant
regurgitation. Cardiac catheterization is indicated to assess the
degree of systemic AV valve regurgitation, to rule out pulmo-
nary hypertension and to assess the systemic right ventricular
function potentially prior to consideration for cardiac trans-
plantation later in life (22).
Figure 20.10 AP projection demonstrating an external conduit
Ebstein’s Anomaly Fontan pathway around the right atrium and directly connecting to
This lesion is due to a lack of delamination of the tricuspid the pulmonary arteries.
valve, resulting in different degrees of TV regurgitation. There
[24/6/010/13:3:7] [242–252]
Figure 20.11 Fenestration within the Fontan baffle (A) before and (B) after percutaneous closure with a 4 mm Amplatzer vascular plug II.
Fontan Pathway Obstruction bilateral Glenn shunts. If these angiograms do not demon-
There are multiple locations for Fontan pathway obstruction. strated the explanation for the patients systemic hypoxemia
These include the distal Fontan anastomosis along the hepatic or embolic events we recommend agitated saline injections in
veins, the proximal anastomosis at the pulmonary arteries due the proximal right and left pulmonary arteries with simultane-
to the Fontan pathway or the SVC anastomosis, and bilateral ous TEE or chest wall echocardiography to assess for tiny
branch pulmonary artery stenoses. Also, the left innominate arteriovenous malformations in either lung. The lung with the
vein can become narrowed especially if there is a history of least or no blood from including the hepatic veins is most likely
upper extremity central line placement during previous proce- to have the malformations. Transcatheter closure (25) of the
dures. All of these locations are amendable to stent intervention venovenous fistulae or larger arteriovenous malformations can
using peripheral size 10 to 22 mm diameter stents. Hemody- be performed using Gianturco coils, Amplatzer vascular plugs,
namic derangements are assessed using mean pressures and or the Amplatzer PDA occluder (Fig. 20.12). Final angiography
careful pullbacks as only a 2- to 3-mmHg gradient can be or agitated saline injections can be utilized to assess for imme-
significant especially if there are venovenous “pop-off” routes diate closure and systemic oxygen saturations at rest or during
for the blood flow. follow-up exercise testing can be checked to confirm improve-
ment and future risk for embolic events.
Persistent Fenestration
This is fairly unusual in the adult population but many youn- Aortopulmonary Collaterals
ger teens and the rare adult may require transcatheter closure A significantly decreased pulmonary blood flow is a stimulus
of the fenestration if there is concern regarding systemic to the development of collateral vessels from the systemic
hypoxemia with or without exercise and if there is a concern circulation. Major aortopulmonary collaterals (MAPCAs) are
regarding possible or previous cerebrovascular events. Multi- present in patients with pulmonary atresia and ventricular
ple different septal occluder devices have been successfully septal defect.
used to close the Fontan fenestration and few data exist using
the covered stent technology to essentially isolate the fenestra-
tion within the Fontan baffle (Fig. 20.11). SPECIAL ISSUES
Pulmonary Hypertension in Congenital
Heart Disease
Venovenous Fistula Changes in the pulmonary vasculature are common in patients
As with the patient with a persistent fenestration the venove- with congenital heart diseases and can be related to increased
nous fistulae will manifest itself as systemic hypoxemia, with or blood flow secondary to left-to-right shunting, distortion of the
without exercise, or a route for systemic embolic events. The pulmonary arteries due to shunts or others (26). Pulmonary
standard angiograms to assess the Fontan circuit for these angiography is frequently performed. The objectives are to rule
malformations is a biplane cineangiogram in the inferior vena out pulmonary branch stenosis (proximal, distal, bilateral, or
cava distal to the hepatic veins, a biplane cineangiogram at the unilateral), assess PCWP, and evaluate response of pulmonary
proximal anastomosis of the Fontan, a biplane cineangiogram pressures to vasodilator tests. Catheterization will help in the
in the left innominate vein for the patient with a right-sided decision-making process of selection for heart or heart and lung
Glenn shunt and angiograms in both SVC in those patients with transplantation in selected patients.
[24/6/010/13:3:7] [242–252]
Figure 20.12 A patient after the Fontan operation with a venovenous fistula from the superior vena cava to the right pulmonary veins
(A) prior to intervention and (B) after percutaneous placement of an Amplatzer vascular plug II.
Secundum atrial septal defects with elevated pulmonary pulmonary valve regurgitation. If the RV fails, signs of right-
vascular resistance or left ventricular diastolic dysfunction sided heart failure will be present, with worsening tricuspid
should be carefully evaluated before proceeding with device valve regurgitation.
closure. Transient balloon occlusion of the defect can be per- Patients are advised to avoid dehydration, heavy exer-
formed to assess the changes in cardiac output and left atrial tion, or systemic vasodilators that can increase the right-to-left
pressure. shunting. If a surgical procedure is planned, careful anesthetic
In patients with D-TGA with atrial switch procedures management (cardiac anesthesia) should be available, and use
(Mustard or Senning) and pulmonary hypertension, pulmonary of an air filter in all intravenous access to avoid paradoxical air
venous baffle obstruction must be ruled out. embolism is mandatory.
When a catheter is wedged in a pulmonary vein, the Avoidance of hypotension is important; otherwise, the
pressure may reflect the pulmonary artery pressure. In cases degree of right-to-left shunting will increase and progressive
that the pulmonary artery pressure cannot be measured, for hypoxemia will develop, with the risk of death. If coronary
example, in a patient with pulmonary valve atresia, pulmonary angiography is needed, the most experienced operator should
vein wedge pressure is used. perform the procedure with minimal contrast to minimize the
risk of kidney failure. Cyanotic patients are more susceptible to
developing nephropathy with the use of contrast, NSAIDs, or
Eisenmenger Syndrome other drugs such as aminoglycosides (27).
Without treatment, the increased pulmonary blood flow result-
ing from a left-to-right shunt will produce progressive struc-
tural changes in the pulmonary vasculature (26). These changes CONCLUSIONS
will consist of medial thickening and hypertrophy, endothelial Congenital heart disease in the adult population, more than any
damage, and in situ thrombosis, resulting in an increase in other type of malformation, is taxing the current medical sys-
pulmonary vascular resistance due to the decrease in the cross- tem as the number and complexity of patient issues is out-
sectional area of the pulmonary circulation and vasoconstric- growing the limited facilities established to care for them. With
tion. As the pulmonary pressures continue to increase, the better survival from childhood, it is evident that few of our
degree of left-to-right shunt will diminish, and eventually surgical and catheter-based interventions are curative, and that
there will be right-to-left shunting, resulting in cyanosis. Eisen- a lifetime of care will be required. The education of both
menger syndrome refers to reversal of a left-to-right shunt to a internal medicine/cardiology trainees as well as those already
right-to-left shunt due to the development of pulmonary vas- in practice, in addition to the development of national data-
cular disease. Patients can present with syncope, cyanosis, bases for tracking patients and their responses to therapy, are
palpitation, hyperviscosity symptoms, hemoptysis, stroke, or critical to the future care of this complex and diverse popula-
brain abscess. tion. Clinical systems designed for the transition of these
The diagnosis is based on physical examination, which patients out of the Pediatric clinics, where many continue to
will disclose clubbing, cyanosis, a right parasternal heave, and receive their care, and into organized, regional ACHD centers
loud P2 with a high pitch decrescendo diastolic murmur of of excellence will facilitate this process.
[24/6/010/13:3:7] [242–252]
REFERENCES 15. de DM, Pilla C, Kattenhorn M, et al. Ambulatory blood pressure,

1. Miyague NI, Cardoso SM, Meyer F, et al. Epidemiological study of left ventricular mass, and conduit artery function late after suc-
congenital heart defects in children and adolescents. Arq Bras cessful repair of coarctation of the aorta. J Am Coll Cardiol 2003;
Cardiol 2003; 80:269–278. 41(12):2259–2265.
2. Inglessis I, Landzberg MJ. Interventional catheterization in adult 16. Hokanson JS, Moller JH. Adults with tetralogy of fallot. Cardiol
congenital heart disease. Circulation 2007; 115:1622–1633. Rev 1999; 7:149–155.
3. Khairy P, Harris L, Landzberg MJ, et al. Implantable cardioverter- 17. Bonhoeffer P, BoudjemlineY, Saliba Z, et al. Percutaneous replace-
defibrillators in tetralogy of Fallot. Circulation 2008; 117(3): ment of pulmonary valve in a right-ventricle to pulmonary-artery
363–370. prosthetic conduit with valve dysfunction. Lancet 2000; 356(9239):
4. Campbell M. Natural history of persistent ductus arteriosus. Br 1403–1405.
Heart J 1968; 30:4–13. 18. Van Praagh R, Jung WK. The arterial switch operation in trans-
5. Mesia CI, Moskowitz WB. Coil Occlusion of elderly ductus position of the great arteries: anatomic indications and contra-
arteriosus. Am J Geriatr Cardiol 1999; 8:131–142. indications. Thorac Cardiovasc Surg 1991; 39(suppl 2):138–150.
6. Latson LA, McManus BM, Doer C, et al. Endocarditis risk of the 19. Warnes CA. Transposition of the great arteries. Circulation 2006;
USCI PDA umbrella for transcatheter closure of patent ductus 114(24):2699–2709.
arteriosus. Circulation 1994; 90:2525–2528. 20. Lock JE. The adult with congenital heart disease: cardiac catheter-
7. Connelly MS, Webb GD, Somerville J, et al. Canadian Consensus ization as a therapeutic intervention. J Am Coll Cardiol 1991; 18(2):
Conference on Adult Congenital Heart Disease 1996. Can J Car- 330–331.
diol 1998; 14:395–452. 21. Dragulescu A, Sidibe N, Aubert F, et al. Successful use of covered
8. Brickner ME, Hillis LD, Lange RA. Congenital heart disease in stent to treat superior systemic baffle obstruction and leak after
adults. First of two parts. N Engl J Med 2000; 342:256–263. atrial switch procedure. Pediatr Cardiol 2008; 29(5):954–956; [Epub
9. Fisher RG, Moodie DS, Sterba R, et al. Patent ductus arteriosus in Dec 19, 2007].
adults–long-term follow-up: nonsurgical versus surgical treat- 22. Graham TP, Bernard YD, Mellen BG, et al. Long-term outcome in
ment. J Am Coll Cardiol 1986; 8:280–284. congenitally corrected transposition of the great arteries. J Am
10. Vita JA, Bittl JA, Selwyn AP, et al. Transcatheter closure of a Coll Cardiol 2000; 36:255–261.
calcified patent ductus arteriosus in an elderly man. J Am Coll 23. Brown ML, Dearani JA, Danielson GK, et al. Functional status
Cardiol 1988; 12:1382–1385. after operation for Ebstein anomaly. J Am Coll Cardiol 2008;
11. Latson LA. Nonsurgical treatment of a neonate with pulmonary 52:460–466, doi:10.1016/j.jacc.2008.03.064.
atresia and intact ventricular septum by transcatheter puncture 24. Fiore AC, Turrentine M, Rodefeld M, et al. Fontan operation: a
and balloon dilation of the atretic valve membrane. Am J Cardiol comparison of lateral tunnel with extracardiac conduit. Ann
1991; 68(2):277–279. Thorac Surg 2007; 83:622–630.
12. Kuhn MA, Latson LA. Transcatheter embolization coil closure of 25. Beekman RH III, Shim D, Lloyd TR. Embolization therapy in
patent ductus arteriosus–modified delivery for enhanced control pediatric cardiology. J Interv Cardiol 1995; 8(5):543–556.
during coil positioning. Cathet Cardiovasc Diagn 1995; 36: 26. Deanfield J, Thaulow E, Warnes C, et al. Management of grown up
288–290. congenital heart disease. Task Force on the Management of Grown
13. Schenck MH, O’Laughlin MP, Rokey R, et al. Transcatheter occlu- Up Congenital Heart Disease, European Society of Cardiology;
sion of patent ductus arteriosus in adults. Am J Cardiol 1993; ESC Committee for Practice Guidelines. Eur Heart J 2003; 24(11):
72:591–595. 1035–1084.
14. Bashore TM, Lieberman EB. Aortic/mitral obstruction and coarc- 27. Berman EB, Barst RJ. Eisenmenger syndrome: current manage-
tation of the aorta. Cardiol Clin 1993; 11(4):617–641. ment. Prog Cardiovasc Dis 2002; 45(2):129–138.
[21/6/010/20:53:12] [253–269]
21
Ventriculography and aortography

José G. Dı́ez and James M. Wilson
VENTRICULOGRAPHY technique that can provide accurate spatial and temporal res-
Introduction olution without relying on geometric assumptions is cardio-
Reliance upon radiographic cardiac-chamber imaging has sub- vascular MRI. Therefore, MRI is to be considered the new, true
stantially declined in recent years because of the success of gold standard for measuring LV systolic function.
echocardiography and, more recently, magnetic resonance Left ventriculography can also be performed with digital
imaging (MRI). In 1980, ventriculography was recommended subtraction angiography (DSA), which uses either an intrave-
as a routine part of the evaluation of patients undergoing nous or a low-dose intraventricular contrast agent. Advantages
cardiac catheterization (1). More recently, other methods of over standard radiography include a reduced use of radiation
accurately determining left ventricular (LV) function or the and contrast media, an ability to visualize with very low
aortic anatomy have supplanted cavity angiography to such a concentrations of contrast medium, and an image format that
degree that the techniques and technical details of obtaining can be directly analyzed by means of quantitative techni-
diagnostic images with ventriculography are being lost. ques (4). By enhancing the contrast-to-background signal,
Nonetheless, cavity imaging (also referred to as “cavitog- DSA allows angiography to be done with reduced doses of
raphy” for lack of a better word) remains important because contrast medium. Nichols and coauthors performed a valida-
it allows efficient, timely diagnosis and guides intervention, tion study that measured LV volume and segmental contrac-
particularly for structural heart disease. The LV ejection tion while also comparing the hemodynamic effects of low- and
fraction (LVEF) provides diagnostic and prognostic informa- high-dose contrast injections in 28 patients (5). The group that
tion in patients with known or suspected heart disease. In received the low-dose contrast injections underwent digital left
clinical practice, the LVEF can be determined with any of five ventriculography and received an intraventricular injection of 7
currently available imaging techniques: contrast angiography, mL of contrast medium diluted in saline solution. This step was
echocardiography, radionuclide blood-pool and first-pass followed by conventional cineangiography of the left ventricle,
imaging, electron-beam computed tomography (CT), and using 45 mL of undiluted contrast medium. After injection of
MRI (2). the diluted contrast medium, LV systolic and end-diastolic
In studies comparing different imaging techniques, the pressures did not change significantly, and patients had no
results have suggested that LVEF measurements are not inter- discomfort. LV volumes calculated from digital ventriculo-
changeable (3). Therefore, conclusions and recommendations grams correlated well with volumes calculated from conven-
should be interpreted in the context of locally available techni- tional ventriculograms. Thus, DSA eliminates the hemodynamic
ques. In addition, there are wide variances in the estimation on effects that result from injecting conventional doses of contrast
volumes and LVEF between different techniques, especially medium, thereby permitting the use of left ventriculography
when using echocardiography. The principal reason for this with markedly reduced doses of contrast medium.
variance is the method used to convert two-dimensional (2D) To determine the validity of intravenous DSA–left ventri-
images to three-dimensional (3D) information. Ventriculography culography (IVDSA-LVG) in evaluating the LVEF and regional
typically uses the equations of Sandler and Dodge; the heart is wall motion, Kuribayashi and coauthors compared IVDSA-LVG
assumed to be symmetrical and shaped like a prolate spheroid. using 30 mL of contrast medium with direct left ventriculogra-
However, this assumption is valid only for the normal heart. phy in 18 patients (6). There was a good correlation between the
Similarly, echocardiographic methods are encumbered by geo- two methods (r ¼ 0.877) in determining the LVEF, and 90% of
metric assumptions, which may result in estimates that vary the interpretations of regional wall motion were in agreement
widely from those of ventriculography. Nuclear methods that between the two methods. Intravenous DSA-LVG was useful
are add-ons to perfusion imaging reconstruct the ventricular and accurate in evaluating the LVEF and regional wall motion.
walls and measure the volume and LVEF with Simpson’s rule. The results of this study suggest that this method may be used in
Unfortunately, these methods are hampered by poor spatial patients with impaired LV function to avoid hemodynamic
resolution (implying a larger error range) and difficulty in derangement induced by conventional, direct left ventriculog-
assessing transmural myocardial infarction (where the LV wall raphy using large doses of contrast medium.
is not visible, the result must be imputed, that is, the location of The above-described ventriculography methods and
the infarcted wall during systole must be inferred). techniques are experimental. They use contrast media, and
Count-determinant, multigated acquisition (MUGA) the radiation doses are actually higher than those used in
scanning provides an accurate LVEF but is not reliable for standard angiography. As these diagnostic imaging methods
determining wall motion. Improvements in the efficiency of have evolved, improvements in standard angiography have
medical diagnosis have virtually excluded MUGA from the kept pace. Cavitography allows clinicians to draw upon angio-
evaluation sequence unless other imaging methods are contra- graphic data; these are the data on which they base most of
indicated or unsuccessful. In essence, the only available their decisions in the cardiac catheterization laboratory.
[21/6/010/20:53:12] [253–269]
Anatomic Considerations and Fundamentals should exceed 20 mL/sec so that the aorta and left ventricle are
In imaging any cardiac cavity, the following variables must be opacified. Any catheter smaller than 6 Fr would not be able to
considered: anatomy, systolic function, regurgitant fraction, achieve the flow rate necessary for opacifying both chambers at a
shunting, cavity size, cavity output, heart rate, the maximum normal heart rate. At a maximum pressure of 1000 psi (for which
flow rate of the diagnostic system chosen, and the catheter most systems are rated) in patients with a normal resting heart
position and injection technique. For standard ventriculogra- rate, a 7-Fr diagnostic system is necessary to inject more than
phy, midcavity positioning of the catheter just below the inflow 60 mL of contrast material over the required three-second period.
of the mitral valve is crucial. This position allows mitral inflow Planimetry is more objective than ventriculography for
to carry the injected contrast material forward, opacifying the quantifying LV function. With planimetry, the projection image
apex of the left ventricle. The preferred angiographic view is of the chamber in question is outlined and the area quantified.
usually the 308 right anterior oblique (RAO) projection. To By means of a series of geometric assumptions, this area is used
evaluate ventricular septal defects or obstructions of the LV to derive the LV volume. The first and most important assump-
outflow tract, a 308 to 608 left anterior oblique (LAO) projection tion is that a ventricle that is somewhat triangular in shape in the
with 208 cranial angulation is necessary (Table 21.1). RAO view is an ellipsoid. In a 2D view, something that looks like
An underlying principle of cavitography is that the contrast a football is an ellipse. In 3D, such a structure is referred to as an
volume of the chamber being imaged should reach at least 10% of ellipsoid. To be precise, a football is a prolate spheroid. This
the total contrast volume for the period of imaging in question. assumption is not uniformly valid and is clearly violated in
Because of the volume of contrast needed for aortography and patients with previous myocardial infarction (the population for
ventriculography, a power injector is typically used. Power- whom the ejection fraction estimation is most important) and
injection parameters include the volume of contrast medium to other patient populations in whom diastolic or systolic anatomy
be infused, maximum injection pressure, injection rate, and tim- deviates from normal. This is the principal reason why echocar-
ing of the pressure increase used to minimize catheter movement. diography and MRI are preferred for measuring LV function.
The settings for adequate cavity imaging are condition depen- The volume of an ellipsoid is calculated with the follow-
dent. For example, a single beat stroke volume of 70 mL would be ing formula:
expected in a patient with no cardiac murmur and with pre-
4 D1 D2 L
sumed normal LV function. Therefore, to achieve the 10% thresh- V¼
old over the target of three to five heartbeats, a volume infusion of 3 2 2 2
30 mL is more than sufficient. Assuming that the heart rate is 60 In the 2D representation, the ellipsoid is an ellipse.
to 70 beats per minute (bpm), this volume may be infused over a Therefore, one may measure the area of the ventricle and the
period of three seconds. The maximum allowed pressure will be length of its long axis. This allows one to calculate the short-
influenced by the necessary injection rate and the size of the axis diameter for the imaginary ellipse with the following
catheter used. For example, power injection with a 4-Fr catheter formulas: RAO, D1 ¼ 4A1/pL1 and LAO, D2 ¼ 4A2/pL2.
system mandates the use of higher maximum allowed pressures After the axes are derived, these formulas can be used to
and relatively low infusion rates (7 mL/sec), as well as longer calculate the ventricular volume. Because the ventricle is some
infusion periods (4 seconds) to provide diagnostic-quality images. distance from the detector, it will be magnified. Unless an
Consider a patient suspected of having severe aortic object of a known size is present within the ventricle when
valve insufficiency. Upon injecting the ascending thoracic the image is being captured, some means of correcting for
aorta with contrast medium, one should be able to quantify magnification is required. The correction factor can be calcu-
the severity of aortic valve insufficiency and, if it is severe, lated from this formula: CF ¼ (H P)/H, where H is the
properly opacify the LV and quantify its volume and function. distance from the tube to the detector and P is the distance from
This goal requires rapidly injecting a large volume of contrast the object to the detector. Alternatively, an object of a known
material over a period of three heartbeats, which is the thresh- size may be placed in the field of view and used for calibration.
old for distinguishing moderate from severe aortic valve insuffi- If the object is placed on or behind the patient, use of the
ciency. It may not be possible to complete an adequate study with correction factor is required (Fig. 21.1).
small-diameter catheters. To meet the above-mentioned thresh- The single-plane method of measuring the ventricular
old, let us assume that the LV end-diastolic volume (LVEDV) is volume assumes that the long-axis and the calculated short-axis
300 mL, the stroke volume is 200 mL, and the LVEF is 66%. When diameters are the same in both the RAO and LAO projections.
the regurgitant volume is 140 mL, the contrast injection flow rate This assumption is acceptable. However, in patient populations
Table 21.1 Anatomic Areas of Interest, Catheter Position, and Angulation Views
Evaluation Catheter position Angiographic view
LVEF Mid-LV 308 RAO, 308 LAO
MR Mid-LV 308 RAO
LVOT Mid-LV 308 LAO, 30–608 cranial
VSD Mid-LV 308 LAO, 308 cranial
Aortic valve/ascending aorta 3–20 cm above aortic valve 308 LAO, 30 cranial
Aortic arch Proximal to right subclavian (innominate) artery 458 LAO
Left atrium (ASD) Right upper pulmonary vein 458 LAO, 458 cranial
Right ventricle Mid-RV AP, lateral
Abbreviations: LVEF, left ventricular ejection fraction; LV, left ventricle; RAO, right anterior oblique; LAO, left anterior oblique; MR, mitral
regurgitation; LVOT, left ventricular outflow tract; VSD, ventricular septal defect; ASD, atrial septal defect; RV, right ventricle; AP, anteroposterior.
[21/6/010/20:53:12] [253–269]
VENTRICULOGRAPHY AND AORTOGRAPHY 255
Figure 21.2 Ventriculogram obtained in right anterior oblique 308

Figure 21.1 Ventriculogram obtained in right anterior oblique 308 view, showing a dilated LV chamber in diastole. Abbreviation: LV,
view. LV measurements obtained with available software from left ventricular.
commercial workstations. Abbreviation: LV, left ventricular.
with abnormal LV anatomy, in systole or diastole, this assump- The angled catheter is more appropriately located within the
tion is unreliable and may lead to errors in the calculation of the LV; this position reduces the incidence of PVCs. Pigtail cathe-
LV volume or function. The formula used in this method is ters have an end loop that keeps the end hole (required for wire
advancement) away from the endocardium, and the other holes
V ¼ 0:81 D2 L CF þ 1:9 mL
6 along the distal shaft provide offset jets that help stabilize the
Automated edge-detection algorithms may be used to catheter and reduce recoil (9).
outline the borders of the LV cavity, thereby decreasing intra- Alternative catheters include those with multiple side holes
and interobserver variability and the analysis time (7). In (e.g., the NIH and Eppendorf catheters). As mentioned above, in
addition, new modalities using 1808 rotational LV injections obtaining measurements from angiographic data, a structure of a
and a series of projection images may be used for 3D recon- known size may be placed within the ventricle. This is most
structions and potentially for 4D reconstructions (8). effectively done by using a pigtail catheter with radiopaque
markers (1 cm from front edge to front edge). Additionally, the
Indications Langston catheter (Vascular Solutions, Inc., Minneapolis, Minne-
Left ventriculography is indicated for the assessment of global sota, U.S.) is a dual-lumen 6-Fr to 8-Fr device with end or side
LV function and regional wall-motion abnormalities (Fig. 21.2). holes 8 cm apart. It is available in angled pigtail, multipurpose,
It can also be used to assess the severity of mitral regurgitation and straight configurations. This catheter provides the ability to
and to identify and assess muscular and membranous ventric- obtain precise simultaneous measurements of aortic and LV
ular septal defects. Other indications include proper position- pressures. However, the lumen size limits the maximum contrast
ing of the CardioKinetix Ventricular Partitioning Device infusion rate, and the pigtail catheter often creates a spurious
(CardioKinetix, Redwood, California, U.S.), which isolates the gradient. The multipurpose version of this dual-lumen catheter is
malfunctioning portion of the left ventricle in patients with much easier to use for hemodynamic measurements.
symptoms of heart failure due to ischemic heart disease. Smaller catheter diameters have been used in recent years
to reduce the probability of groin-access complications. How-
Equipment ever, very small catheter diameters may limit maximum injec-
Ventriculography is best performed with an angled pigtail tion rates, impairing the evaluation of the severity of
catheter, which avoids some of the pitfalls of end-hole catheters regurgitant lesions. Small catheter diameters (4 Fr and 5 Fr)
such as inadequate opacification, ectopy, myocardial staining, are usually sufficient for performing coronary angiography, left
and catheter movement (recoil). The straight pigtail catheter ventriculography, and aortography in patients whose LV size
frequently becomes oriented beneath the posterior leaflet of the and function and aortic diameter are normal. However, for the
mitral valve, resulting in poor opacification of the LV or evaluation of dilated chambers or regurgitant lesions, larger
apically produced premature ventricular complexes (PVCs). catheter diameters (6 Fr or 7 Fr) may be necessary.
[21/6/010/20:53:12] [253–269]
In most instances, a standard 0.035-in. J wire is sufficient ascending aorta. Although the proximal connector tubing
for positioning the catheter and crossing the aortic valve. In the allows angiography, we have found that it tends to rupture at
setting of aortic valve stenosis, a smaller, soft-tipped, 0.032-in. high pressures.
straight wire is frequently used. If multiple exchanges are An alternative technique for entering the left ventricle of
necessary after the valve is crossed, a stiff exchange-length patients with acquired aortic valve stenosis uses the right
wire is highly useful. Examples include the J-tipped Amplatz Judkins and Amplatz right-modified catheter and a soft-tipped
(Cook Medical, Bloomington, Indiana, U.S.) or 3-cm tipped 0.032-in. straight wire. The alternative technique may be used
Amplatz devices. for normal patients or aortic valve stenosis. In normal patients,
a standard J-wire may be used. In patients with acquired aortic
valve stenosis, the soft-tipped 0.032-in. straight wire is pre-
Clinical Aspects ferred. The catheter is advanced over the wire to the right
Crossing the Aortic Valve coronary sinus. The straight wire is then placed outside the
Once an indication for performing ventriculography has been catheter and oriented so that it can be advanced to the non-
established, the first step is to cross the aortic valve. As in several coronary cusp. The catheter is rotated clockwise until aortic
other invasive procedures, crossing the aortic valve in patients ejection creates visible wire oscillation. The catheter is then
without significant aortic stenosis is regarded as straightforward. withdrawn by means of the exchange technique, so that the
The presence of significant aortic stenosis necessitates a modified wire is oriented within the aortic valve orifice. The wire may
technique (see later in text). Although bioprosthetic valves may then be advanced into the left ventricle. After the catheter is
be crossed with catheters, it is not advisable for any catheter or inserted into the left ventricle, the 0.035-in. J wire may be used
wire to be placed across a tilting-disc valve prosthesis, because to exchange for the catheter of choice to be used for additional
the equipment can be entrapped. In addition, crossing a mechan- measurements or angiography.
ical prosthetic valve may cause sudden severe aortic regurgita-
tion, leading to hemodynamic complications. Ventriculography Technique
The classic approach for crossing the aortic valve in Hemodynamic measurements should be performed before
patients without stenosis requires the use of a 0.035-in. wire, ventriculography to obtain baseline information.
which is placed in the pigtail catheter and positioned slightly With single-plane equipment, ventriculography will pro-
back from the tip. The catheter is placed above the valve, and is vide only a 2D projection of the ventricle, and individual images
advanced until it prolapses above the valve and forms a loop will not include all the LV segments. The standard views for
resembling a reverse J shape or the number 6. The loop is then ventriculography are the RAO (308), which shows the antero-
pulled back and rotated slightly in a clockwise direction. Dur- lateral, anterior, apical, and inferior ventricular walls, and the
ing systolic opening of the aortic valve, the catheter will fall into LAO 608 or 208 cranial view, which allows better imaging of
the LV outflow tract, it is advanced into the LV cavity. On the lateral, posterior, and septal ventricular walls (Fig. 21.3). The
occasion, the body of the catheter will prolapse across the valve,
but the tip will stay in the aortic root. In this event, advancing
the 0.035-in. wire will usually cause the catheter to prolapse
into the ventricle. This can result in severe ventricular ectopy
and may even damage the valve leaflets. Having the patient
take a breath while the pigtail catheter is advanced into the
ventricle during systole will sometimes facilitate entry.
Aortic valve stenosis poses a more complex situation. This
condition carries a risk of embolization from the valve or from
thrombus that may form on the wire or catheters if multiple
attempts are required, thereby prolonging the procedure. In
these cases, anticoagulation with heparin (3000–5000 units
bolus) is recommended. Because of the heightened risk, crossing
the valve should be done only if noninvasive assessment is
inconclusive or interventions (e.g., valvuloplasty or percutane-
ous prosthetic valve deployment) are planned.
In patients with aortic stenosis, a preliminary aortic root
angiogram will allow visualization of the valve opening. Fluo-
roscopic and cineangiographic guidance of the wire through
the calcified opening can also be done, in both the RAO and
LAO projections. Once the opening has been identified, a
straight-tipped 0.032-in. wire is inserted through an Amplatz
left 1 catheter or a similarly configured device such as an
Amplatz left 2, an Amplatz right 1 modified, a multipurpose
1, or a Feldman catheter. Once the wire crosses the valve, the
diagnostic catheter is advanced, and pressures are recorded. If Figure 21.3 Diagrammatic representation of RAO and LAO views
an angiography catheter or other catheter is required, it is of the LV obtained during contrast angiography showing division of
exchanged over an exchange-length wire (usually a soft-tipped, the LV wall into 10 numbered segments. Abbreviations: RAO, right
shaped wire, because it will be less likely to induce ventricular anterior oblique; LAO, left anterior oblique; LV, left ventricular.
ectopy). With a 6-Fr dual-lumen pigtail catheter, one can Source: From Ref. 10.
simultaneously measure pressures in the left ventricle and the
[21/6/010/20:53:12] [253–269]
Figure 21.4 Ventriculogram obtained in right anterior oblique 308 view with an angled pigtail catheter. (A) Left ventricle in diastole, (B) left
ventricle in systole.
LAO views are particularly useful in patients with lateral ventricular septal defect may be missed due to incomplete
ischemia (especially circumflex ischemia), suspected ventricular opacification.
septal defect, and mitral regurgitation (Fig. 21.4). Some operators
prefer to use biplane ventriculography. Settings
The LV cavity is usually visualized with 30 to 50 mL of Optimal ventriculography is performed with a power injector
contrast material. Some clinical scenarios call for a modification so as to fill the LV cavity. Adjustable settings on the power
of this volume. For example, in patients with known or sus- injector include pressure and flow rates, volume, and the rate of
pected mitral regurgitation, 50 to 60 mL of contrast material is pressure increases. In each case, the settings will vary slightly,
needed to completely opacify the left atrium. In elderly patients
with hypertensive disease and small LV cavity, smaller vol-
umes (30–36 mL) are adequate. If the patient has severe valvu-
lar disease, LV dysfunction, or an elevated end-diastolic
pressure, use of a nonionic contrast agent is recommended. In
patients with a severely elevated LV end-diastolic pressure
(LVEDP), ventriculography should be performed only after
pharmacologic interventions (e.g., nitroglycerin or furosemide
administration) have been carried out to prevent subsequent
pulmonary edema (Fig. 21.5).
Once advanced into the left ventricle, the pigtail catheter
should be positioned in the midventricular cavity over the
region of mitral inflow. Apical positioning of the catheter
might lead to excessive ectopy, which could interfere with the
interpretation of wall-motion abnormalities. A position that is
too basal may interfere with the mitral apparatus, leading to an
overestimation of mitral regurgitation. A catheter that is difficult
to manipulate, ectopically positioned, or visibly entangled may
indicate that the mitral apparatus is involved. In this setting,
proper catheter positioning usually requires catheter withdrawal
(preferably over a wire) and replacement into the left ventricle.
Repositioning the catheter usually requires countering
and pulling it (ideally over the wire) and then advancing it once
the tip is free (Fig. 21.6).
If the patient is in unstable condition and DSA is avail- Figure 21.5 Ventriculogram showing severe mitral regurgitation,
able, manual injection may be attempted while the patient which was classified as 4þ on angiographic assessment. Arrows
holds his or her breath. This method should be used only for demonstrate the left atrium.
a quick estimate, because important abnormalities such as a
[21/6/010/20:53:12] [253–269]
lumen) with respect to the quality of the resulting cine left

ventriculograms (11). Ventricular tachycardia and couplets
occurred at similar frequencies in the groups tested. Patients
in the angled pigtail subgroups had significantly more frequent
catheter-induced mitral regurgitation. Nevertheless, the inter-
group opacification and overall quality of the left ventriculo-
grams were similar. The patients with marginal opacification
were significantly heavier (P ¼ 0.004), with larger LVEDVs (P ¼
0.019), and smaller contrast volumes per EDV (P ¼ 0.005). The
patients with excellent left ventriculograms were significantly
younger (P ¼ 0.019), were more frequently females (P ¼ 0.036),
and tended to be less heavy (P ¼ 0.09). Ventricular tachycardia
was the most common cause of unsatisfactory left ventriculo-
grams. In the RAO view, deeper (more apical) placement of the
catheter was associated with a heightened incidence of ven-
tricular tachycardia (53%). The most “silent” area in terms of
catheter-induced arrhythmias was the posterobasal region.
With 6-Fr high-flow pigtail catheters and nonionic contrast
agents, a more basal catheter position and a higher contrast
volume will increase the quality of the left ventriculograms.
Low-volume ventriculography has been shown to be
similar to standard, larger-volume ventriculography for the
estimation of LV systolic function. In 102 patients, Hodges
and coworkers compared low-volume ventriculography to
Figure 21.6 Ventriculogram from a patient with known LV hyper- standard-volume ventriculography using standard (15 mL/
trophy. The angled pigtail catheter was maintained in the midcavity sec for 3 seconds) and low-volume (15 mL/sec for 1 second)
without compromising the mitral valve apparatus (no mitral regurgi- contrast agents (12). Each patient served as his or her own
tation). Abbreviation: LV, left ventricular. control. Of the 204 ventriculograms, 27% were not interpretable
because of ectopy. Ectopy involving 3 beats was more com-
mon with standard-volume angiograms (41% vs. 14%; P <
0.001). With both methods of contrast-agent injection, the
LVEDP increased from baseline levels (P < 0.001). In patients
for whom both angiograms could be interpreted (n ¼ 58), no
based on patient sex, ventricular size, and catheter type and differences were noted between LVEFs measured with planim-
size. Generally, injection of 30 to 40 mL at a flow rate of 10 to etry (low-volume method ¼ 61 20% vs. standard-volume
15 mL/sec is sufficient for imaging a normal ventricle at a method ¼ 62 20%; r ¼ 0.87; P < 0.001). Therefore, low-
normal heart rate. A rate of pressure rise of 0.5 to 1.0 second is volume ventriculography reduces the contrast load and ectopy
needed to prevent lunging of the catheter, which can lead to while providing estimates of the LVEF similar to those obtained
increased ectopy. The pressure rate settings are typically 400 to with standard volumes.
600 psi, 600 to 900 psi, and 800 to 1200 psi for a 6-Fr, 5-Fr, and 4-
Fr system, respectively. However, in our experience, adequate Quantification of Ventricular Function
visualization has been obtained with a 5-Fr catheter at 400 psi Depending on the angiographic projection, specific ventricular
and with a 4-Fr catheter at 600 psi. It is useful to test the wall-motion abnormalities may be identified. These include
integrity of the system when using high pressures and a 4-Fr hypokinesia (decreased but not absent motion of a ventricular
system. Although most manifolds would tolerate these high segment); akinesia (a complete absence of wall motion); tardi-
pressures, short connectors might rupture. To prevent a cata- kinesia (delayed contraction of a ventricular segment); and
strophic embolism during ventriculography, one should pay dyskinesia (paradoxical expansion or wall motion, usually
careful attention to the connections and remove all air from the due to tethering from the adjacent segments). The LVEF calcu-
injector system and the catheter before using it. lated with ventriculography will usually be slightly higher than
During injection, the operator’s left hand should be kept that calculated with echocardiography.
on the catheter to change its position as required during the The American College of Cardiology (ACC), American
procedure. If the catheter is too far into the cardiac apex, the Heart Association (AHA), and other cardiovascular organiza-
operator should pull it back and reposition it in the midcavity. tions recognize the importance of having clinical data stan-
This is particularly useful to decrease arrhythmias. Conversely, dards that define and standardize platforms and conditions.
if the catheter tends to go back into the aorta, it should be These elements are described in the ACC/AHA’s 2008 publi-
advanced. cation titled “Key Elements and Definitions for Cardiac Imag-
Ventricular arrhythmias or inadequate opacification of ing” (10). The authors acknowledge that each imaging modality
the ventricular cavity during contrast-cineangiographic left uses a unique range of values for quantitatively determining
ventriculography frequently interfere with evaluation of the LVEF and that, even within a single modality, different
regional-wall motion, the LVEF, or mitral regurgitation. In a quantitative methods may yield disparate results. The consen-
prospective, randomized study of 95 patients, Deligonul and sus is that systolic function should be classified into four
associates compared a traditional (straight) pigtail catheter to a categories: normal, mildly reduced, moderately reduced, and
large-loop pigtail catheter (both 6-Fr devices with a large severely reduced. The quantitative value may be reported as a
[21/6/010/20:53:12] [253–269]
specific value (e.g., 64%) or as a 5% range (e.g., 30–35%). The Table 21.2 Angiographic Assessment of Mitral Regurgitation
midpoint of the range would be used for data collection and
Grade Explanation
storage. Regional function of myocardial segments assessed by
means of contrast LV angiography should be described as 1þ Brief and incomplete atrial opacification over several
normal, hypokinetic, akinetic, dyskinetic, or not visualized. cycles. Clears rapidly. No atrial enlargement.
The authors also suggest the use of a 10-segment division in 2þ Moderate opacification of the atria with each cycle.
Never greater than LV opacification. No significant
the RAO or LAO views. In the RAO view, the segments should
LA enlargement.
be described as anterobasal, anterolateral, apical, diaphrag- 3þ Atrial opacification equal to ventricular opacification.
matic, and posterobasal. In the LAO view, they should be Delayed clearing of atria over several cycles.
described as basal septal, apical septal, posterolateral, infero- Significant enlargement of the LA.
lateral, and superior lateral (Fig. 21.3). 4þ Atrial opacification in less than 3 cycles and greater
than that of the ventricle. Severe enlargement of
Quantification of Mitral Regurgitation the LA. Opacification of the pulmonary veins.
Valvular regurgitation affects the cardiac chambers upstream Abbreviations: LV, left ventricle; LA, left atrium.
or downstream from the leaking valve. The ability of these
chambers to accommodate the excess workload is highly vari-
able and at least partly related to the time span over which the Table 21.3 Regurgitant Fraction Estimation in Mitral Regurgitation
regurgitation has developed. For example, if the left atrium has
Regurgitant fraction Equivalent
had insufficient time to dilate after rupture of a chorda tendi-
nea, that chamber’s compliance produces striking increases in 20% Grade 1þ regurgitation described visually
the left atrial pressure in response to the volume introduced 21–40% Grade 2þ regurgitation
from the left ventricle. Conversely, in chronic, slowly progres- 41–60% Grade 3þ regurgitation
60% Grade 4þ regurgitation
sive regurgitation, the compliance characteristics of the left
atrium are altered so that the volume of blood ejected into
the left atrium may be accommodated with little change in
the effective forward flow (EFF). The following formula can be
pressure. As a result, the pressure measured within the accept-
used to calculate the EFF:
ing chamber may be a poor measure of the severity of regur-
gitation. Moreover, regurgitant severity may not be the sole EFF ¼ ð1 RFÞ EF EDV
determinant of the hemodynamic effect of valvular failure.
The same formula may be used to quantify the regurgi-
Angiographic quantization of valvular regurgitation
tant volume if the SV and net cardiac output are known. As a
severity depends on opacification of the proximal chamber
general rule, when the EFF begins to decrease because the
during contrast injection. Therefore, the visual measures that
chamber has enlarged and cannot accommodate the excess
are commonly used depend on the severity of regurgitation, as
load, the valve needs to be repaired.
well as the volume of the accepting chamber and the volume of
The difference between the angiographic SV and the
flow through the affected chamber. High-volume flow may
forward SV is the regurgitant volume. The angiographic SV is
limit opacification and result in rapid clearance, and low-
computed from the left ventriculogram, and the forward SV is
volume flow may do the opposite. A severely dilated chamber
derived from the cardiac output, as determined by the Fick or
may be relatively difficult to opacify sufficiently to reach a
thermodilution method and the heart rate. The RF is the portion
threshold for assigning maximal severity. Thus, angiographic
of the angiographic SV that does not contribute to the net
methods are accurately characterized as “semiquantitative.”
cardiac output. It is computed as the regurgitant SV divided
Valvular regurgitation may be readily visualized angiograph-
by the angiographic SV (Table 21.3).
ically. This requires sufficient contrast administration for excel-
lent opacification of the distal chamber and relies on the degree
of opacification of the proximal chamber and the time required Limitations
for the contrast agent to clear. Administration of 40 to 60 mL of Ventriculography has been compared with other methods and
contrast agent over three seconds is generally necessary. technologies for assessing LV function and morphology. In
Both the RAO and the LAO cranial projections can be used 65 consecutive patients, Takenaka and associates compared
to identify significant mitral regurgitation. Grading the amount 2D echocardiography, thermodilution techniques, and biplane
of regurgitation is based on the degree of opacification of the cineventriculography with respect to LVEDV, LV end-systolic
atrium and ventricle, as well as the atrial size, and the number of volume (LVESV), SV, and LVEF (calculated with the modified
cycles required for maximal opacification (Table 21.2). Both Simpson rule). 2D echocardiography was performed within
elevation of the left atrial pressure in acute regurgitation and three days of cardiac catheterization (13). The results were
dilation of the left atrium in chronic regurgitation can interfere compared with those obtained by means of the thermodilution
with the use of this grading system. technique and biplane cineventriculography. The heart rate and
SV were significantly different among the three techniques, and
Regurgitant Fraction ventriculography yielded the highest values. These findings
Perhaps the most valuable indicator of regurgitation severity is suggest that patients may have been in a hyperadrenergic state
the regurgitant fraction (RF). The regurgitant volume is the caused by anxiety during invasive cineventriculography and
amount of ejected blood that is returned to the chamber in thermodilution examinations. The inter- and intraobserver
question during the period of preparation for the next beat. The variabilities for echocardiography differed little from the
RF is the regurgitant volume divided by the stroke volume variability for ventriculography. Although there were good
(SV). The effect of regurgitation may be estimated by combin- correlations between the echocardiographic and cineventriculo-
ing the RF and LVEF with the LVEDV, which gives an idea of graphic findings for the LVEDV (r ¼ 0.67), LVESV (r ¼ 0.80),
[21/6/010/20:53:12] [253–269]
and LVEF (r ¼ 0.78), as assessed by two independent observers, Contraindications

there was a lack of agreement for the LVEDV, LVESV, and Contraindications for left ventriculography include critical left
LVEF. The echocardiographic LVEDV values were significantly main disease, a tilting-disc aortic prosthesis (the catheter can
lower than the cineventriculographic values. cause acute aortic regurgitation, become entrapped, or damage
3D echocardiography, being exempt from the need for the tilting mechanism), decompensated heart and/or renal
geometric assumptions, correlates highly with ventriculogra- failure, and a recently diagnosed (freshly formed) intracardiac
phy for estimating ventricular volumes. The former technique thrombus. Because thrombi more than six months old may
has approximately half the variability of 2D echocardiography have a lower risk of dislodgement, some operators may pro-
in making these measurements (14). ceed with ventriculography in their presence. However,
Radionuclide ventriculography is a valuable tool in the because of the uncertainty involved in appropriately estimating
risk stratification of postinfarct patients (15), but in this patient the age of an intracavitary thrombus, we suggest deferring
population, LVEF correlates poorly with radiographic ventri- ventriculography and proceeding with noninvasive evaluation.
culography (r ¼ 0.42). Therefore, in this setting an alternative
method is warranted. Nonfluoroscopic electromechanical map-
Complications
ping of the left ventricle is feasible and safe. The LVESVs
Among the complications that have been encountered during
obtained with this method strongly correlate with those mea-
ventriculography are ventricular arrhythmias, embolization of
sured by means of ventriculography, but the LVEF does
air or thrombus, contrast agent–related complications, worsen-
not (16).
ing of hemodynamic values in patients with decompensated
The utility of ventriculography in providing an estima-
heart failure, and myocardial staining or “tattooing.”
tion of the LVEF that is reproducible and correlate with sur-
Air or thrombus embolization has been observed during
vival has been well demonstrated. Many LVEF calculations
ventriculography and left-sided heart catheterization. Trans-
used for prognosis in the literature are, in fact, angiographic.
cranial Doppler (TCD) imaging has shown that these compli-
However, it should be clear that LVEF values obtained with
cations commonly occur when patients are evaluated with
alternative methods cannot be “interchanged” with values
ventriculography. To assess the prevalence, time of occurrence,
obtained by means of ventriculography.
and potential significance of microembolic signals (MESs)
detected with TCD during left-heart catheterization, Leclercq
and coworkers monitored the right and left middle cerebral
Special Issues (Complications and Other
artery in 51 consecutive patients (36 men, 15 women) (21).
Cardiac Cavities)
MESs were detected in all except two patients (mean number of
The following relative contraindications are shared by cardiac
signals, 17.1 12.8 per patient) mainly during left ventriculog-
catheterization and angiography: severe uncontrolled hyper-
raphy. The MESs were asymptomatic and probably of gaseous
tension, ventricular arrhythmias, acute stroke, severe anemia,
origin because they occurred predominantly during contrast
active gastrointestinal bleeding, allergy to radiographic contrast
media injection and were not related to the patients’ cardio-
agents, acute renal failure, decompensated congestive heart
vascular history or to atheroma risk factors. No neurologic
failure (the patient cannot lie flat), unexplained febrile illness
events occurred within 24 hours after ventriculography in the
and/or untreated active infection, electrolyte abnormalities
49 patients who had MESs.
(e.g., hypokalemia), and severe coagulopathy. Some of these
An unusual but potentially dangerous complication of
factors can be corrected before the procedure, thereby lowering
left-sided heart catheterization is massive air embolization
the risk.
(22,23). Therefore, the system should be meticulously checked
In patients with cardiogenic shock undergoing percuta-
for air before performing the injection. If air bubbles embolize
neous revascularization, contrast ventriculography is deferred
to the brain or coronary arteries, they can cause transient or
until after coronary angiography has been performed (17) or is
permanent neurologic complications and hemodynamic col-
avoided to prevent further hemodynamic compromise result-
lapse, respectively. The operator should be alerted if a sensory
ing from arrhythmias or increases in ventricular pressure and
deficit is noted after a contrast medium has been injected for
renal failure. Contrast medium–induced hypotension and bra-
left ventriculography. Theoretically, the preferred treatment for
dycardia were once serious concerns in patients with severe
cerebral air embolization is hyperbaric oxygen. However, in
aortic stenosis and left main or severe three-vessel coronary
clinical practice this is rarely performed.
artery disease, but these complications have been decreased by
the use of low-osmolar or nonionic media (18). In cases of
severe coronary artery disease (including left main) disease, it Left Atrial Angiography
is uncertain whether the risks of a left ventriculogram are The left atrial anatomy is complex, and the chamber borders are
outweighed by its benefits (19). However, as discussed above, poorly defined, making assessment of left atrial function and
ventriculography can be safely performed with less than 40 mL volume somewhat difficult. However, opacification of the left
of a low-osmolar contrast agent. Proper access technique and atrium may be useful for the anatomic evaluation of an atrial
careful manipulation of the catheter are required because septal defect (ASD) or a patent foramen ovale (PFO). Similarly,
hemodynamic instability can be caused by a catheter-induced it may be important to evaluate the left atrial anatomy before
arrhythmia or vasovagal response. and after radiofrequency ablation procedures. In most instan-
Left ventriculography should be performed only in ces, the anatomy may be assessed quite well with noninvasive
patients in whom the benefits from the information obtained means. In fact, due to the complex and highly variable anatomy
are greater than the risks related to the procedure. Otherwise, of the pulmonary veins, MRI and CT are superior to angio-
noninvasive assessment of ventricular function and mitral graphic projection imaging. However, angiographic guidance is
regurgitation by means of echocardiography and nuclear or necessary for placement of ASD, PFO, or left atrial occluder
MRI techniques is more appropriate (20). devices. Left atrial angiography is best performed with the NIH
[21/6/010/20:53:12] [253–269]
catheter, which is placed in the right upper pulmonary vein. As new technologies and interventions become available,
This allows contrast material to “wash” along the interatrial such as percutaneous aortic valve replacement, it is important
septum. A 458 LAO imaging plane with 458 cranial angulation that cardiologists become skilled at crossing the aortic valve
best outlines the septum to show ASDs and PFOs. Injection and appropriately assessing LV function and morphology.
parameters for left atrial angiography are similar to those for Meanwhile, cavitography remains an important tool for diag-
ventriculography, except that lower maximum pressures may nostic evaluation and for planning interventional procedures.
be used in the low-pressure left atrium. Its principles are straightforward, but close attention to detail is
necessary to obtain diagnostic-quality images.
Right Ventriculography
The right ventricle’s shape is poorly conducive to evaluation by
means of projection imaging. It is both triangular in the ante- AORTOGRAPHY
roposterior projection and discoid in the lateral projection. Introduction
Therefore, use of angiographic imaging for volume estimates Aortography is the radiographic technique used for opacifying
is fraught with error, and the geometric assumptions used in the lumen of the aorta, the superior aspect of the aortic valve
imaging the right ventricle are of little value. However, right leaflets, and all of the vessels that arise from the aorta (24).
ventricular (RV) angiography may be useful for evaluating Noninvasive radiographic evaluation of the aorta and its
tricuspid regurgitation, global RV function, suspected RV car- branches has evolved rapidly because of advancements in
diomyopathy, or arrhythmogenic RV dysplasia, and for plan- imaging techniques such as computed tomography angiogra-
ning a pulmonary valvuloplasty. The severity of tricuspid phy (CTA) and magnetic resonance angiography (MRA). Nev-
regurgitation is evaluated qualitatively, and the grading system ertheless, catheter-based angiographic evaluation remains an
is similar to that used for mitral regurgitation: 1þ, trivial; 2þ, integral part of the diagnostic process and the main guide to
mild; 3þ, moderate; and 4þ, severe. Given that access to the choosing an intervention (either endovascular or surgical). An
right ventricle requires crossing of the tricuspid valve, evalua- ascending aortogram allows one to determine the competency
tion of tricuspid insufficiency is as potentially erroneous as of the aortic valve, the anatomy and diameter of the ascending
imaging of the right ventricle. thoracic aorta, and the presence of patent aortocoronary bypass
Ideally, right ventriculography is best performed with a grafts. When the catheter is positioned proximal to the inno-
biplane system. When performed with a single-plane system, it minate artery, the examiner may evaluate the anatomy of the
is usually done in the AP projection. With a biplane system, the aortic arch and its major arterial branches, detect the presence
608 LAO projection is added. The contrast volume ranges from of a persistent ductus arteriosus or coarctation, and gain infor-
20 to 40 mL, injected over two seconds. In general, lower mation about the descending thoracic aorta. When the catheter
maximum pressures may be used when RV pressures are low. is positioned more distally, one may evaluate aneurysms,
Either an angled pigtail catheter, a variant of the angled dissections, and abnormalities of the abdominal aorta and its
pigtail catheter known as the Grollman catheter, or the NIH arterial branches (Fig. 21.7).
catheter may be used for this procedure. The authors prefer the
NIH catheter because it is easily maneuverable in patients who
have a difficult anatomy, and it can be concurrently used for
subselective pulmonary artery angiography. The catheter is
placed in the midcavity position, and injection parameters
similar to those of left ventriculography are used. When RV
pressures are elevated, pulmonary artery catheterization
should be performed before right ventriculography. In the
setting of severe pulmonary hypertension, main pulmonary
angiography or right ventriculography may result in hemody-
namic compromise, so alternative methods for assessing RV
anatomy and function should be chosen.
Conclusions: Ventriculography
Catheterization of the left ventricle to obtain hemodynamic
measurements and to visualize the left ventricle with contrast
ventriculography is an important component of a complete
angiographic study because it provides essential anatomic
and functional information. Although noninvasive testing tech-
niques continue to improve and become more accurate, cardiac
catheterization remains the standard for evaluating hemody-
namic parameters. In some instances, catheterization is pre-
ferred (e.g., when patients present acutely with a myocardial
infarction). By assessing myocardial and valvular function with Figure 21.7 Aortogram in 608 left anterior oblique projection from a
ventriculography, the clinician may quickly gain information patient who had previously undergone repair of an ascending aortic
vital to making sound decisions about a patient’s acute care. aneurysm with a Dacron graft (white arrows indicate the beginning
In obese patients who have suboptimal echocardiographic and end of the graft). Aortic dilatation distal to the graft compromises
windows, ventriculography may provide information not the arch and descending aorta (black arrows).
obtainable from the echocardiogram.
[21/6/010/20:53:12] [253–269]
Anatomic Considerations and Fundamentals

During aortography, variants in the anatomy may be identified.
To understand and appropriately identify these variants, some
basic concepts need to be reviewed. In the earliest stages of
embryogenesis, the vessels are plexiform. As the fetus grows,
the vessels become recognizable as conduits. The adult aorta
and the aortic arch system result from the regression and fusion
of six pairs of aortic arches (Table 21.4). Normal regression of
the right dorsal aortic root results in the normal left aortic arch.
Aortic arch branch variants can be identified in 30% to 35% of
individuals. The most frequent variant (20%) is a common
origin of the innominate and left common carotid arteries
(bovine arch). Other variants include a left vertebral artery
arising from the arch (5%), an aberrant right subclavian artery
or lusoria (1%), and, less frequently, a right-sided aortic arch,
double aortic arch, or cervical aortic arch (Fig. 21.8).
The abdominal aorta is formed from the right and left
dorsal aortas. The numerous splanchnic branches that supply
the primitive digestive tract are reduced to become the celiac,
superior, and inferior mesenteric arteries. Anomalies of the
abdominal aorta itself are rare, but they are common in the
primary branches (Fig. 21.9) (25). Multiple renal arteries can be
present in a third of the population. Approximately 2% to 7% of
patients have bilateral accessory renal arteries. Persistence of
the ventral connection results in the rare condition known as Figure 21.8 Digital subtraction angiogram from a woman with a
celiomesenteric trunk. In addition, the hepatic, left hypogastric, history of dysphagia who presented with right-arm claudication. The
or splenic arteries may have separate origins from the aorta. occluded, aberrant right subclavian artery originates distal to the left
The presence of multiple branches accounts for the parietal subclavian artery. Collateral flow is reconstituting the aberrant ves-
sel (arrow).
collateral systems that can be identified in occlusive diseases of
the aorta. (Intercostal, subcostal, and lumbar arteries can pro-
vide collaterals to the iliolumbar and superior gluteal branches
of the internal iliac artery and to the circumflex branches of the
external iliac artery.) The visceral pathways for collaterals arise
from the superior and inferior mesenteric arteries.
Even in healthy humans, the aorta lengthens with age,
primarily because the ascending aorta elongates over time (26).
This elongation is also a function of blood pressure. At the same
time, an increase in collagen augments aortic wall stiffness.
Because these changes are balanced by changes in body mor-
phology, the changes in the physical characteristics of the aorta
cannot be detected by measuring the pulse wave velocity.
Aged, hypertensive patients will frequently have a tortuous
aorta that challenges the interventionalist with regard to cath-
eter passage, catheter placement, and proper angiographic
imaging. Because of the difficulties that the abdominal
aorta may present when it is severely diseased (e.g., with an
Table 21.4 Anatomic Result of Aortic Arch Regression and Fusion

During Embryogenesis
Vessel Origin
Proximal ascending aorta Truncus arteriosus
Distal ascending aorta, innominate artery, Aortic sac
and arch until left common carotid artery
Right subclavian artery Fourth aortic arch
Common carotid arteries Third aortic arch
Aortic arch between the left common carotid Fourth aortic arch Figure 21.9 Abdominal aortogram from a patient with diffuse
artery and the left subclavian artery atherosclerosis, dyslipidemia, and hypertension who presented
Left subclavian artery Left intersegmental with claudication. This image, obtained in the anteroposterior
artery view, shows bilateral renal artery stenosis (white arrows) and severe
Ductus arteriosus Sixth aortic arch distal aortic stenosis (black arrow).
Descending aorta Left dorsal aorta
[21/6/010/20:53:12] [253–269]
aneurysm, a dissection, a luminal thrombus, or tortuosity), a

radial or brachial approach may be needed during aortography
to obtain adequate images yet avoid complications.
Optimal aortography requires the use of a power injector
to opacify the aorta adequately. In each case, a slight variation
in settings will be required, based on the size of the aortic root,
presence of aneurysms, degree of aortic valve insufficiency,
diameter of the catheter lumen, and patient size. The catheters
used for aortography are similar in shape to those used for
ventriculography (pigtail catheters), but, because diameter
measurements are required, a marker pigtail catheter is pre-
ferred. Alternatively, other angiographic catheters such as the
Omniflush or the tennis racket can be used.
Indications
Although aortography has been supplanted by transesophageal
echocardiography (TEE), CTA, and MRA for many disorders of
the aorta, angiographic imaging may be useful to exclude aortic
dissection (Fig. 21.10) in patients referred for urgent coronary
angiography. This technique is also useful for measuring the
aortic diameter—preferably with a marker pigtail—when other
studies have rendered conflicting information, for measuring
the severity of aortic insufficiency, for delineating the aortic and
coronary anatomy before surgical repair of an aneurysm, and Figure 21.10 Aortogram in left anterior oblique demonstrating an
for verifying the presence of an aortoenteric fistula (27,28). ascending aortic dissection. The dissection extends from the
Perhaps the most important advance in aortography in sinuses of Valsalva to the origin of the innominate artery (arrows).
recent years has been the use of digital subtraction, which
allows anatomic details to be accurately delineated with rela-
tively small quantities of contrast media. Digital subtraction Contrast aortography has gained a central role in guiding
angiography is superior to standard imaging for evaluating aortic endovascular aneurysm repair (EVAR) and other percu-
congenital and acquired lesions of the arch and great vessel taneous interventional procedures at the level of the aorta
origins (29). (Fig. 21.11) and the supra-aortic vessels.
Figure 21.11 Aortogram from a critically ill patient with multiple comorbidities who presented with a large pseudoaneurysm several months
after undergoing surgical repair of an ascending aortic aneurysm. (A) Evidence of a perforation at the level of the distal suture in the Dacron
graft (arrow). (B) Successful closure of the graft perforation with an Amplatzer1 Cribriform Occluder (diameter, 18 mm) AGA Medical
Corporation, Plymouth, Minnesota, U.S. (off-label indication) (arrow).
[21/6/010/20:53:12] [253–269]
Equipment
For contrast aortography, angiographic equipment that pro-
vides a spatial resolution of at least 3 to 4 lines/mm is required.
Ideally, large image intensifiers up to 16 in. (40.6 cm) in diam-
eter should be used to provide a field of view large enough to
show the entire arch and its branches or the entire abdominal
aorta at the same time. Current technologies include postpro-
cessing and image-reformatting modalities, which allow diam-
eter analysis, 3D rendering, and rotational 3D rendering. Flat-
panel technology has significantly increased image resolution.
For arterial access, 4-Fr to 7-Fr systems can be used.
Dedicated radial sheaths or micropuncture devices (usually
4–6 Fr) are used for radial or brachial access. The length of the
guidewire is determined by its intended use and varies from
145 to 175 cm. However, if multiple catheter exchanges are
required, or if maintaining position is important (i.e., in the
presence of aortic dissections, complex aneurysms, severe tor-
tuosity, or a large atherothrombotic burden), a long (260–350 cm)
exchange wire is used. While the guidewire’s diameter may
range from 0.014 to 0.038 in., most of the time a 0.035-in.
guidewire is used.
Angiographic catheters are usually pigtail, marker pigtail,
Omniflush, or tennis racket devices or catheters with multiple
side holes. A power injector is required for optimal opacification. Figure 21.12 Angiogram obtained via common femoral access in a
man with a history of an ascending thoracic aortic aneurysm. The
Low-osmolar or iso-osmolar contrast agents are preferred because
complex dissection flap (arrows) prohibited passage of the wires or
they cause less intravascular volume augmentation, fewer side
catheters into the ascending aorta. Subsequent ascending aortic
effects, and possibly less contrast-induced nephropathy. angiography and coronary angiography were performed via the right
radial approach.
Clinical Aspects
In general, careful planning for the procedure is required. The
access site needs to be determined first. Choosing the appro-
priate site of access for peripheral arteriography is a most
important procedural decision, which is analogous to planning
a surgical incision (30). The site of access is determined on the
basis of anatomic knowledge acquired from previous noninva-
sive studies (if available), the patient’s clinical history (includ-
ing ascending or abdominal aortic pathology), and the physical
examination (including the presence of pulses). In our practice,
the most common sites of access are the common femoral (70%)
(Fig. 21.12) and radial (30%) (Fig. 21.13) arteries. With the aid of
fluoroscopic guidance, access via the common femoral artery is
usually obtained with a micropuncture system, which is later
exchanged for a catheter with a larger French size.
In most of our angiographic studies, a 5-Fr system is
used. Also, we most commonly use exchange-length wires of
0.035 in., including the Wholey Hi-Torque model (Mallinckrodt
Inc., St. Louis, Missouri, U.S.), the Bentsen model (Cook, Inc.,
Bloomington, Indiana, U.S.), and the hydrophilic Glidewire
(Terumo, Somerset, New Jersey, U.S.). If diameter measure-
ments are important, a marker pigtail catheter is used.
When we are dealing with a complex anatomy (involving
an aortic dissection, previous aneurysm repair with a Dacron
graft, or severe tortuosity) and are uncertain about reaching the
aortic root via the femoral approach, we usually obtain access
with a micropuncture system. We advance a 0.035-in. Wholey
wire until it reaches the ascending aorta. After establishing that
the wire is in the true lumen (by verifying that the wire moves Figure 21.13 Angiogram performed via the right radial route with a
freely, responds to flow from the aortic valve, and loops easily marked pigtail showing an ascending aortic aneurysm (white
at the sinuses), we exchange the micropuncture system for a 5- arrows). Note the dissection flap at the level of the aortic arch
Fr system (including a long sheath if necessary) and then (black arrow). There is a Dacron graft extending from the prosthetic
advance the angiographic catheters. Severe aortic tortuosity aortic valve to the mid-segment of the ascending aorta, after which
can cause difficulties in advancing and manipulating the cath- the dissection can be visualized.
eter. In such cases, a long access sheath may help when one is
[21/6/010/20:53:12] [253–269]
over an hydrophilic wire (e.g., Glidewire). At all times, careful

manipulation of wires and catheters is essential to prevent
traumatic disruption of intraluminal thrombus and/or wall
atheromas.
Ascending Aortography
When the ascending aorta is visualized in the 308 LAO projec-
tion with 308 cranial angulation, dissections of the ascending
aorta, anomalous coronary origins, and some saphenous bypass
grafts may best be seen (Fig. 21.16). In contrast, the aortic arch
and branch origins are best displayed in the 458 LAO projection
with no cranial angulation. The catheter (preferably a regular or
marker pigtail) is placed in the ascending thoracic aorta at the
level of the sinotubular ridge (2–3 cm above the aortic valve).
This allows the catheter to stiffen and lengthen during power
injection without encountering the aortic valve. Unintentional
contact with the aortic valve may either produce iatrogenic
aortic valve insufficiency or result in poor opacification of two
of the three sinuses of Valsalva. If the catheter is positioned just
proximal to the right innominate branch, the aortic arch and the
origin of the great vessels can be seen. The 308 RAO projection is
best reserved for visualizing saphenous vein grafts, but it may
also be used to demonstrate aortic insufficiency.
Figure 21.14 Aortogram performed via a common femoral Technique

approach, showing an ascending aortic aneurysm that extends A 4-Fr to 7-Fr pigtail catheter is generally used to perform
into the arch. aortography. The size is critically dependent on the specific
situation in question. To assess the aortic anatomy, the injection
velocity from a 4-Fr catheter is sufficient with the assistance of
DSA. For aortic insufficiency, a catheter with a larger diameter
working from a femoral approach (Fig. 21.14). However, one (7 Fr) will allow rapid contrast injection. However, any catheter
should carefully consider using a radial or brachial approach. without end holes may be used. With end-hole catheters, there
With the radial approach (Fig. 21.15), we prefer to use a may be a risk of aortic dissection or aortic valve damage during
dedicated radial sheath (Terumo) and to perform exchanges power injection (Figs. 21.17 and 21.18).
Figure 21.15 (A) Aortograms in anteroposterior projection from a 65-year-old man with a history of hypertension. The descending aorta was
completely interrupted below the level of the left subclavian artery. A small collateral vessel is seen (arrow). (B) Aortogram obtained via the
left radial approach, showing aortic coarctation, with collaterals reconstituting the descending aorta.
[21/6/010/20:53:12] [253–269]
Figure 21.16 Aortogram of the ascending aorta (left anterior Figure 21.18 Aortogram showing dilatation of the sinuses of
oblique 308 view with cranial angulation) showing a bicuspid aortic Valsalva and the presence of aortic regurgitation (grade 2þ).
valve and immediate filling of the left ventricle allowing the diagnosis
of severe aortic insufficiency (grade 4þ).
straight as possible within the constraints placed on it by the

anatomy of the aorta. To reduce this “kick” during injection, the
rate of pressure rise during power injection is modulated. A
period of pressure rise of 0.5 to 1.0 second is often used, the rise
time being a function of the size of the catheter lumen. The
pressure settings are typically 600, 900, and 1200 psi for a 6-, 5-,
and 4-Fr system, respectively. Generally, injection of 40 to 50 mL
at 20 to 25 mL/sec is sufficient for imaging a normal aorta. To
prevent a catastrophic embolism during aortography—particu-
larly if performed at the level of the ascending aorta or aortic
arch—careful attention must be paid to removing air from the
injector system before it is used. According to anecdotal case
reports, air injection of 50 cc in the ascending aorta has proven
fatal.
Quantification of Aortic Insufficiency

Quantification of aortic valve regurgitation requires injection of
a sufficient volume of contrast material (about 60 mL) into the
aorta via a catheter that is near, but not in contact with, the
aortic valve. The severity of regurgitation is rated according to
a semiquantitative scale similar to the one used for mitral
regurgitation (Table 21.5).
Figure 21.17 Aortogram from a patient who had undergone aortic Abdominal Aortography
valve replacement with a single-tilt disc valve and repair of an Abdominal aortography is usually performed via the common
ascending aortic aneurysm with a Dacron graft several years earlier. femoral approach with a multi-side-hole catheter (4–6 Fr).
Note the severe dilatation of the aortic root (arrows).
Depending on the patient’s history and physical findings,
radial or brachial access may be preferred. Once access has
been obtained, a soft-tip wire is placed across the area of
Settings interest with the aid of fluoroscopy, and the angiographic
During injection, the high-viscosity contrast agent encounters catheter is advanced so that its tip is at the level of the T12 or
resistance along the length of the catheter lumen. This resistance L1 vertebra. If available, a biplane system allows the visceral
produces a force within the catheter, making the device as branches to be visualized with only one injection of contrast
[21/6/010/20:53:12] [253–269]
Table 21.5 Angiographic Assessment of Aortic Insufficiency contrast agents during diagnostic aortography or EVAR may
reduce the incidence of renal dysfunction after the proce-
Grade Explanation
dure (31). Carbon dioxide angiography is a safe alternative to
1þ Brief and incomplete ventricular opacification. Clears the use of iodinated contrast media and is vastly underused.
rapidly. When conducted through the endograft delivery sheath, carbon
2þ Moderate opacification of the ventricle that clears in less dioxide angiography is not only reliable for endograft deploy-
than 2 cycles. Not greater than aortic root opacification.
ment but also safe, nontoxic, and inexpensive. In addition, it
3þ Opacification of the ventricle equal to aortic root
may expedite EVAR by eliminating the need for a number of
opacification in more than 3 cycles. Delayed clearing
of ventricle over several cycles. angiographic catheter placements and exchanges during the
4þ Opacification of the ventricle almost immediately or in procedure.
less than 3 cycles that is equal to or greater than that Conventional angiography is widely regarded as the gold
of the aortic root with delayed clearing of the ventricle. standard for classifying endoleaks. Recently, with the advent of
time-resolved MRA (TR-MRA), faster magnetic resonance gra-
dients have allowed rapid data acquisition and review of
vascular data on a real-time continuous angiogram (32). The
initial results show that TR-MRA is an effective noninvasive
method for classifying endoleaks. It may allow screening of
patients with endoleaks to identify leaks that require urgent
repair.
In peripheral arterial disease, 3D dynamic contrast-
enhanced subtraction MRA has high sensitivity and specificity.
In segments with more than mild stenosis, MRA is 97.1%
sensitive and 99.2% specific (33). This technique is a noninva-
sive alternative to conventional angiography for screening
patients suspected of having lower-extremity ischemia.
CTA reveals mural changes in addition to luminal abnor-
malities (34). Those abnormalities are high-attenuation wall in
precontrast transverse CT images, circumferential wall thicken-
ing with or without enhancement, a concentric low-attenuation
ring inside the aortic wall, and mural enhancement in delayed
phase. CTA may provide valuable information regarding dis-
ease activity and progression.
Special Issues
A severely diseased aorta can present multiple technical
difficulties during angiography, resulting in potential compli-
cations. A difficult anatomy may necessitate excessive
manipulation of catheters, causing distal embolization of athe-
rothrombotic debris. In the presence of aortic wall dissections,
Figure 21.19 Abdominal aortogram from a 68-year-old woman the operator may not realize that the catheter is in the false
with a history of Takayasu’s arteritis, showing bilateral renal artery lumen; these dissections can be extended by the catheter,
stenosis (arrows) and a significant decrease in the diameter of the
causing organ perfusion compromise and perforations.
distal abdominal aorta.
During the procedure, wires, catheters, and other devices
may penetrate the aortic wall, initiating dissection and/or
rupture. The International Registry of Aortic Dissection (35)
indicates that up to 5% of acute aortic dissections are iatrogenic
(27% being due to percutaneous procedures). Patients with
material. The celiac (level T12-L1), superior mesenteric (L1-L2), iatrogenic dissections are more likely to have myocardial
and inferior mesenteric (L3-L4) arteries have an anterior origin, ischemia (36% vs. 5%; P < 0.001) or a myocardial infarction
while the renal arteries originate laterally, just below the origin (15% vs. 3%; P < 0.001) than are patients with noniatrogenic
of the superior mesenteric artery. The required settings are dissections. The former group also has a higher mortality
similar to those used for imaging the ascending aorta and arch. (35% vs. 24%). In addition, when the ascending aorta and
The abdominal aorta bifurcates into the common iliac arteries at arch are manipulated, stroke can occur with devastating con-
about the L4 level (Fig. 21.19). sequences (36,37).
Atheroembolism can also lead to renal insufficiency,
which develops slowly over several weeks. Other systemic
Limitations manifestations of atheroembolism include livedo reticularis,
The limitations of aortography include the use of contrast abdominal or foot pain, and purple toes associated with sys-
agents, which can result in nephrotoxicity; the use of radiation, temic eosinophilia (blue toe syndrome) (38). Systemic compli-
which can be harmful especially to younger patients; and the cations may also occur. They are related to allergic and
use of arterial access, which can lead to complications such anaphylactoid reactions in 3% of cases (<1% of which involve
as embolization of atherothrombi. Avoidance of nephrotoxic hospitalization) (39).
[21/6/010/20:53:12] [253–269]
Conclusions: Aortography Evaluation of left ventricular volume and ejection fraction. Int J
The multiple noninvasive imaging modalities that have Card Imaging 1995; 11:255–262.
evolved for evaluating the aorta and its branches include 14. Sapin PM, Schroder KM, Gopal AS, et al. Comparison of two- and
MRA, CTA, and duplex ultrasonography. These methods three-dimensional echocardiography with cineventriculography
for measurement of left ventricular volume in patients. J Am
allow clinicians to appropriately diagnose patients, follow up
Coll Cardiol 1994; 24:1054–1063.
their conditions, and plan appropriate therapeutic strategies 15. Urena PE, Lamas GA, Mitchell G, et al. Ejection fraction by
while avoiding the complications inherent in invasive percuta- radionuclide ventriculography and contrast left ventriculogram.
neous procedures. Aortography still plays a primary role in the A tale of two techniques. SAVE Investigators. Survival and
invasive evaluation of aortic disorders (valvular, traumatic, ventricular enlargement. J Am Coll Cardiol 1999; 33:180–185.
aneurysmatic, and atherothrombotic). Before elective surgical 16. Van Langenhove G, Hamburger JN, Smits PC, et al. Evaluation of
repair of thoracic aortic aneurysms, aortography and coronary left ventricular volumes and ejection fraction with a nonfluoro-
angiography provide important information about the relation- scopic endoventricular three-dimensional mapping technique.
ship of nearby vessels to the aneurysm and about coronary Am Heart J 2000; 140:596–602.
artery locations and patency. Aortography is also fundamental 17. O’Neill WW. Interventional therapy of cardiogenic shock. In:
Stack RS, Roubin SR, O’Neill WW, eds. Interventional Cardiovas-
in guiding therapeutic endovascular interventions. Careful
cular Medicine, Principles and Practice. 2nd ed. Philadelphia, PA:
procedural technique is required to maximize benefits and Churchill Livingstone, 2002:363–379.
minimize the risk of complications. 18. Barrett BJ, Parfrey PS, Vavasour HM, et al. A comparison of
nonionic, low-osmolality radiocontrast agents with ionic, high-
REFERENCES osmolality agents during cardiac catheterization. N Engl J Med
1. Hood WP, Rackley CE, Grossman W. Cardiac ventriculography. 1992; 326:431–436.
In: Grossman W, ed. Cardiac Catheterization and Angiography. 19. Bergelson BA, Tommaso CL. Left main coronary artery disease:
2nd ed. Philadelphia, PA: Lea & Febiger, 1980:170–184. assessment, diagnosis, and therapy. Am Heart J 1995; 129:350–359.
2. Rumberger JA, Behrenbeck T, Bell MR, et al. Determination of 20. Keeley EC, Ellis SG, Grines CL. Left main coronary interventions.
ventricular ejection fraction: a comparison of available imaging In: Stack RS, Roubin SR, O’Neill WW, eds. Interventional Cardio-
methods. The Cardiovascular Imaging Working Group. Mayo vascular Medicine, Principles and Practice. 2nd ed. Philadelphia,
Clinic Proc 1997; 72:860–870. PA: Churchill Livingstone, 2002:635–646.
3. Bellenger NG, Burgess MI, Ray SG, et al. Comparison of left 21. Leclercq F, Kassnasrallah S, Cesari JB, et al. Transcranial Doppler
ventricular ejection fraction and volumes in heart failure by detection of cerebral microemboli during left heart catheterization.
echocardiography, radionuclide ventriculography and cardiovas- Cerebrovasc Dis 2001; 12:59–65.
cular magnetic resonance; are they interchangeable? Eur Heart J 22. Goldenberg I, Shupak A, Shoshani O, et al. Left ventriculography
2000; 21:1387–1396. complicated by cerebral air embolism. Cathet Cardiovasc Diagn
4. Vogel RA. Left ventricular imaging by digital subtraction angiog- 1995; 35:331–334.
raphy. Int J Card Imaging 1988; 3:29–38. 23. Kearney KR, Smith MD, Xie GY, et al. Massive air embolus to the
5. Nichols AB, Martin EC, Fles TP, et al. Validation of the angio- left ventricle: diagnosis and monitoring by serial echocardiogra-
graphic accuracy of digital left ventriculography. Am J Cardiol phy. J Am Soc Echocardiogr 1997; 10:982–987.
1983; 51:224–230. 24. Delany DJ. Aortography. In: Grossman W, ed. Cardiac Catheter-
6. Kuribayashi S, Ootaki M, Matsuyama S, et al. [Evaluation of ization and Angiography. 2nd ed. Philadelphia, PA: Lea &
left ventricular function by digital subtraction angiography: effect Febiger, 1980:197–211.
of dose and administration mode of contrast medium, and 25. Valji K. Abdominal aorta. In: Valji K, ed. Vascular and Interven-
comparison with direct ventriculography]. J Cardiogr 1985; 15: tional Radiology. Philadelphia, PA: W B Saunders, 1999:102–126.
575–584. 26. Sugawara J, Hayashi K, Yokoi T. Age-associated elongation of the
7. Oost E, Oemrawsingh P, Reiber JH, et al. Automated left ven- ascending aorta in adults. J Am Coll Cardiol Imgaging 2008; 1:
tricular delineation in X-ray angiograms: a validation study. Cath- 739–748.
eter Cardiovasc Interv 2009; 73:231–240. 27. Gundry SR, Williams S, Burney RE, et al. Indications for aortog-
8. Manzke R. Intra-operatory volume imaging of the left atrium and raphy in blunt thoracic trauma: a reassessment. J Trauma 1982; 22:
pulmonary veins with rotational X-ray angiography. Med Image 664–671.
Comput Assist Interv 2006; 9:604–611. 28. Kram HB, Wohlmuth DA, Appel PL, et al. Clinical and radio-
9. Baim DS. Cardiac ventriculography. In: Baim DS, ed. Grossman’s graphic indications for aortography in blunt chest trauma. J Vasc
Cardiac Catheterization, Angiography and Intervention. 7th ed. Surg 1987; 6:168–176.
Philadelphia, PA: Lippincott, Williams & Wilkins, 2006:222–226. 29. Chernin MM, Pond GD, Sahn DJ. Digital subtraction angiography
10. Hendel RC, Budoff MJ, Cardella JF, et al. ACC/AHA/ACR/ASE/ of the aortic arch. Cardiovasc Intervent Radiol 1984; 7:196–203.
ASNC/HRS/NASCI/RSNA/SAIP/SCAI/SCCT/SCMR/SIR 2008 30. Jaff MR, MacNeill BD, Rosenfield K. Angiography of the aorta and
Key data elements and definitions for cardiac imaging: a report of peripheral arteries. In: Baim DS, ed. Grossman’s Cardiac Catheter-
the American College of Cardiology/American Heart Association ization. 7th ed. Philadelphia, PA: Lippincott, Williams, & Wilkins,
Task Force on Clinical Data Standards (Writing Committee to 2006:254–282.
Develop Clinical Data Standards for Cardiac Imaging). J Am Coll 31. Criado E, Kabbani L, Cho K. Catheter-less angiography for
Cardiol 2009; 53:91–124. endovascular aortic aneurysm repair: a new application of carbon
11. Deligonul U, Jones S, Shurmur S, et al. Contrast cine left ventri- dioxide as a contrast agent. J Vasc Surg 2008; 48:527–534.
culography: comparison of two pigtail catheter shapes and anal- 32. Lookstein RA, Goldman J, Pukin L, et al. Time-resolved magnetic
ysis of factors determining the final quality. Cathet Cardiovasc resonance angiography as a noninvasive method to characterize
Diagn 1996; 37:428–433. endoleaks: initial results compared with conventional angiogra-
12. Hodges MC, Rollefson WA, Sample SA, et al. Comparison of low- phy. J Vasc Surg 2004; 39:27–33.
volume versus standard-volume left ventriculography. Catheter 33. Sueyoshi E, Sakamoto I, Matsuoka Y, et al. Aortoiliac and lower
Cardiovasc Interv 2001; 52:314–319. extremity arteries: comparison of three-dimensional dynamic con-
13. Takenaka A, Iwase M, Sobue T, et al. The discrepancy between trast-enhanced subtraction MR angiography and conventional
echocardiography, cineventriculography and thermodilution. angiography. Radiology 1999; 210:683–688.
[21/6/010/20:53:12] [253–269]
34. Park JH. Conventional and CT angiographic diagnosis of 38. Rosman HS, Davis TP, Reddy D, et al. Cholesterol embolization:
Takayasu arteritis. Int J Cardiol 1996; 54(suppl):S165–S171. clinical findings and implications. J Am Coll Cardiol 1990; 15:
35. Januzzi JL, Sabatine MS, Eagle KA, et al. Iatrogenic aortic dissec- 1296–1299.
tion. Am J Cardiol 2002; 89:623–626. 39. Bettmann MA, Heeren T, Greenfield A, et al. Adverse events with
36. Armstrong PJ, Han DC, Baxter JA, et al. Complication rates of radiographic contrast agents: results of the SCVIR Contrast Agent
percutaneous brachial artery access in peripheral vascular angiog- Registry. Radiology 1997; 203:611–620.
raphy. Ann Vasc Surg 2003; 17:107–110.
37. Fayed AM, White CJ, Ramee SR, et al. Carotid and cerebral
angiography performed by cardiologists: cerebrovascular compli-
cations. Catheter Cardiovasc Interv 2002; 55:277–280.
[2/7/010/8:45:24] [270–274]
22
Pulmonary angiography
Syed Sohail Ali and P. Michael Grossman
INTRODUCTION setting is right ventricular failure. Patients may present with

Even though the diagnostic imaging modalities for the evalu- wide range of symptoms and, therefore, a high index of suspi-
ation of pulmonary vasculature have made considerable cion is necessary for the prompt diagnosis. Symptoms may
advancements over past couple of decades, pulmonary angiog- include dyspnea, hemoptysis, cough, and pleuritic chest pain.
raphy still remains the gold standard technique for the diag- Syncope indicates a hemodynamic compromise. Clinical signs
nosis of pulmonary embolism (PE). However, because of more include elevated jugular venous pressure, decreased pulmonic
widely available noninvasive imaging techniques such as mul- component of second heart sound, right ventricular heave and
tislice computed tomography (CT) and magnetic resonance tachycardia—all indicating right ventricular dysfunction. Wells
imaging (MRI), the utilization of pulmonary angiography as a and coworkers have described clinical means to estimate the
primary diagnostic modality has declined considerably. Cath- probability of deep venous thrombosis and PE (3,4).
eter angiography is now more commonly used for various Electrocardiography (ECG) in acute PE usually shows
endovascular interventions on the pulmonary circulation, sinus tachycardia. The presence of classic S1Q3T3 pattern may
such as mechanical embolectomy, embolization for tumors or help in making diagnosis; however, this is not commonly seen.
retrieval of foreign bodies, as well as for the diagnosis and Other findings may include incomplete or complete right bun-
treatment of a variety of congenital heart diseases. dle branch block and right axis deviation. The presence of Qr
pattern in lead V1 and inverted T waves in anterior precordial
leads indicates increased risk of poor clinical outcomes (5). A
ANATOMIC CONSIDERATIONS negative D-dimer has high negative predictive value (>90%)
Pulmonary Circulation Anatomy and low specificity (45%) for PE (6–8). Therefore, the test is
The main pulmonary artery originates from the right ventricle useful only as a “rule-out” modality in the office setting or
and travels anteriorly on the left side of the aorta and then emergency room. Arterial blood gas analysis should not be
follows a posterior course, bifurcating into the right and left used for screening purposes because of it’s low specificity in
pulmonary arteries (Fig. 22.1). The right pulmonary artery PE, but may help direct therapy. The presence of hypoxemia,
gives rise to right upper-lobe branch during its course in hypocapnia, respiratory alkalosis, and increased alveolar-arte-
mediastinum that further divides into three upper-lobe seg- rial gradient is usually seen (9).
mental arteries. The right pulmonary artery continues and then Chest CT is now considered initial imaging test for the
divides into middle-lobe and lower-lobe segmental arteries. diagnosis of PE. Chest radiographs and ventilation perfusion
The left pulmonary artery continues in the mediastinum and lung scanning should precede pulmonary angiography and
gives rise to variable number of small segmental arteries to the serve as a planning aid. The ventilation perfusion scan is very
upper lobe. It then bifurcates into interlobaris and basalis useful for diagnosis of PE when the chest radiograph is normal
branches that give rise to two lingular and four lower-lobe and CT chest with intravenous contrast cannot be per-
segmental arteries, respectively. Further branching of these formed (10). A normal or a high probability lung scan are
vessels is remarkably variable. The segmental pulmonary diagnostic of PE. To interpret low or intermediate probability
veins are also variable; however, they form superior and infe- scans, clinical history should be considered to help direct any
rior vein on each side before draining into the left atrium. further use of diagnostic modalities and management.
The sensitivity of multidetector CT ranges from 70% to
FUNDAMENTALS 90% (11,12). MR imaging can also be utilized for diagnosis of
Patients referred for pulmonary angiography almost invariably PE and can reach sensitivity and specificity as high as pulmo-
have acutely or chronically elevated right heart pressures and nary angiography. MR angiography is more time demanding
may be hemodynamically unstable. Therefore, pulmonary and less available in the acute setting than CT angiography. In
angiograms should be performed by experienced operators in addition, highly symptopmatic patients may not comply suffi-
laboratories with staff and equipment capable of invasive ciently to have adequate quality images with MR. Venous
hemodynamic monitoring. ultrasonography can be performed in patients with symptoms
of deep venous thrombosis and, if a thrombosis is confirmed,
the finding is sufficient to diagnose PE if the patient has a
INDICATIONS suggestive clinical presentation. Contrast venography is rarely
Pulmonary Embolism performed except when catheter-directed thrombolysis or other
PE is one of the important causes of cardiovascular morbidity percutaneous intervention are planned, or an inferior vena cava
and mortality with an annual incidence of 1 per 1000 in the adult filter is to be inserted. Transthoracic echocardiography is an
general population (1,2). The main cause of death in the acute important tool for risk stratification in PE. Accordingly, right
[2/7/010/8:45:24] [270–274]
PULMONARY ANGIOGRAPHY 271
for surgical pulmonary thromboembolectomy. Pulmonary

hypertension may persist after thromboembolectomy and iden-
tifies patients with poor outcomes. Lung transplantation may
be an option in patients with extensive chronic embolic disease.
Primary pulmonary hypertension is a disease of unknown
etiology. Angiography in such patients shows nonspecific dila-
tation of the proximal arteries with smooth, rapid tapering of
distal vessels with corkscrew appearance of the distal arteries. If
untreated, the right ventricle fails secondary to gradual increase
in pulmonary vascular resistance. Vasodilator challenge with IV
adenosine or inhaled nitric oxide is often performed before
initiation of vasodilator therapies. Reduction of 20% in pul-
monary vascular resistance with decrease in mean pulmonary
artery pressure 20% is considered a positive response. A
Figure 22.1 Pulmonary arteries. Right pulmonary artery (A) (i ) right vasodilator challenge may also be performed to document
upper-lobe segmental arteries, (ii ) middle-lobe segmental arteries, reversibility in pulmonary vascular resistance in a patient
(iii ) lower-lobe segmental arteries. Left pulmonary artery (B) (i ) left being considered for heart transplantation.
upper-lobe segmental arteries; (ii ) lower-lobe segmental arteries; Pulmonary arteriovenous malformations are usually
(iii ) lingular segmental arteries. asymptomatic. However, patients may present with dyspnea,
hemoptysis, cyanosis, and clubbing. Polycythemia and para-
doxical embolism may occur due to extracardiac right-to-left
shunting. A contrast-enhanced spiral chest CT can be utilized
ventricle dysfunction is associated with increased mortality and to establish the diagnosis, and pulmonary angiography is usu-
in the presence of hemodynamic instability warrants thrombol- ally performed only to guide percutaneous embolization.
ysis, catheter intervention, or surgical embolectomy. Pulmonary artery stenosis is usually seen in patients who
Predictors of a positive angiogram include the following: survive repair of congenital heart diseases such as tetralogy of
Fallot, truncus arteriosus, pulmonary valvular stenosis, patent
1. Presence of one or more risk factors for pulmonary throm- ductus arteriosus, aortic stenosis, ventricular septal defects, or
boembolism (bed rest, surgery, or trauma within 6 weeks, transposition of great vessels during childhood. Pulmonary
previous history of deep vein thrombosis or PE, malignancy, angiography is both diagnostic and part of therapeutic proce-
congestive heart failure); dures in which such lesions can be treated with angioplasty
2. Chest radiograph showing infiltrates, pleural effusion, or and stents.
atelectasis; Pulmonary artery aneurysms—usually secondary to pul-
3. Ventilation perfusion scan interpreted as indeterminate or monary hypertension—are seen in patients who underwent
high probability. corrective surgery for congenital heart disease; in infectious
When only one of the three predictors is present, the diseases like tuberculosis and syphilis; and in rheumatologic
likelihood of angiographically demonstrable PE is <5%, with processes such as Behcet disease. Percutaneous embolization
two of three predictors positive, about 20% and with all three can be performed after pulmonary angiography.
predictors being positive, the likelihood is >70%. Partial anomalous pulmonary venous return can be diag-
The natural history of pulmonary thromboembolism sug- nosed by pulmonary angiography with oxygen saturation run
gests that most patients will have identifiable emboli for at least and delayed filming to identify the venous phase and to
one week following the acute episode. However, the number of quantify left-to-right shunting.
false-negative studies will increase after 48 hours and the per-
formance of angiography after 7 days may not be advisable.
EQUPIMENT AND PROCEDURE
Venous Access
Other Indications The right common femoral vein is the most often utilized access
Pulmonary angiography and right heart catheterization can be site for pulmonary angiography due to the ease of cannulation
utilized to both confirm the diagnosis and for the evaluation of and of compression, and its relatively straight course into right
the possibility of surgical treatment in a patient with pulmo- side of the heart through the inferior vena cava. The left
nary hypertension secondary to chronic thromboembolic dis- common femoral vein is used if the right side is not accessible
ease. The characteristic angiographic findings of chronic for reasons such as proximal deep venous thrombosis, fibrosis,
thromboemboli include pouching of contrast in organized infection, or recent hematoma. If iliac or inferior vena cava
thrombus with obstruction of the distal artery, the presence thrombosis is suspected at the time of the procedure, 15 mL of
of thin or thick webs or bands that appears as radiolucencies in contrast can be injected into the femoral vein to confirm or rule
lobar or segmental vessels and can cause stenosis with or out the presumptive diagnosis.
without poststenotic dilatation. Luminal irregularities are com- The internal jugular vein can be utilized if no femoral
mon with vessel tapering of large pulmonary arteries and approach is possible. The basilic vein at the antecubital fossa is
obstruction of lobar arteries, usually at their origin. Such preferred over the internal jugular vein if thrombolytic therapy
findings were confirmed surgically in one study of 250 is being considered because it allows for a better hemostasis at
patients (13). In addition to chronic anticoagulation, placement the puncture site. The cephalic vein may not be a good choice
of an inferior vena cava filter should be considered. Patients due to its relatively smaller size and presence of acute angle
with proximal pulmonary artery emboli should be considered when it joins the axillary vein.
[2/7/010/8:45:24] [270–274]
Choice of Contrast and Injection Rates

Traditionally, low-osmolar contrast agents are used for pulmo-
nary angiography. The advantage of such agents includes less
frequent side effects of flushing, hypotension, nausea, or cough
reflex. Currently, there is limited data regarding iso-osmolar,
nonionic agents for use in pulmonary angiography. The determi-
nants of contrast injection rate include the rate of blood flow in the
catheterized vessel, pulmonary artery pressure, the type of cath-
eter used for angiography, and the imaging mode. Digital imag-
ing techniques require less contrast to be injected for adequate
opacification of selected arteries. The injection rate and volume
should be less if more subselective catheterization is performed in
smaller vessels. This should also be considered in patients with
pulmonary hypertension and right ventricular overload to avoid
hemodynamic side effects (14). Right ventricular end-diastolic
pressure of 20 mmHg is usually considered the upper limit for
safe use of contrast media in chronic pulmonary hypertension (15).
If available, biplane angiography may also be used to further
reduce both total contrast volume and catheter dwell time.
Imaging Modes
Digital subtraction angiography (DSA) is equivalent to conven-
tional angiography in image quality and diagnostic perfor-
mance (16). In one study, comparable accurate detection of PE
was noted by use of DSA as compared to conventional angiog- Figure 22.2 Catheters used in pulmonary angiography. (A)
raphy (17). There are many advantages of DSA as the imaging Berman angiographic balloon catheter (Arrow International, Inc.,
mode. It allows the use of less contrast media, which is partic- Reading, Pennsylvania, U.S.); (B) Van Aman pulmonary pigtail
ularly important in patients with pulmonary hypertension or catheter (Cook, Inc., Bloomington, Indiana, U.S.); (C) Montefiore
renal insufficiency. The images are acquired rapidly in flexible catheter (Cook, Inc., Bloomington, Indiana, U.S.).
display format, in which images can be viewed individually or in
cine format for both DSA and conventional angiography.
The main disadvantage of DSA is requirement of image provides safety during right heart catheterization as well as
acquisition without motion that may be suboptimal in patients provides easy access to any segmental artery (Fig. 22.2). The
with severe cardiac or pulmonary symptoms, and in patients retrieval of such catheters should, however, always be made
who are unable to breath hold. Motion artifacts can be reduced after straightening with a guidewire under direct fluoroscopic
by mask-shifting techniques that may improve cardiac motion; visualization to avoid entrapment in the tricuspid valve or
however, it is less helpful in respiratory motion artifacts. For subvalvular apparatus. The balloon-tipped catheters have side
that reason, DSA is usually not possible in patients undergoing holes in the shaft of the catheter proximal to the balloon and
urgent angiography for massive PE other than if the patient is power injection after balloon occlusion allows for selective
intubated, a situation in which DSA should be performed after high-quality injections. After deflation of the balloon, the cath-
induction of an apnea. eter can be removed without the use of fluoroscopy.
The two standard views that have been validated in a In general, contrast medium is injected separately into
large clinical trial are the anteroposterior and the 458 ipsilateral right and left pulmonary arteries rather than the main pulmo-
oblique view (18). Lateral views are not useful because the nary artery, with an injection rate per artery of approximately
frequently observed reflux of the contrast into the opposite 20 mL/sec for total of 30 to 40 mL depending on the patient size
lung may hinder interpretation. at a maximum pressure of 600 psi. The regions that are
suspected to be abnormal are examined first in anterior poste-
Technique rior or oblique planes or with biplane angiography. Further
Pulmonary angiography is usually performed in the acute injections may be required to better define the anatomy.
setting for evaluation and catheter-directed treatment of mas-
sive PE. Close invasive blood pressure and electrocardio- CLINICAL ASPECTS
graphic monitoring is required for early detection of A normal pulmonary angiogram is demonstrated in Figure 22.3.
hypotension, brady- or tachyarrhythmias, and atrioventricular Often selective or superselective injections with magnification
block during the procedure. Angiography is initiated by per- may be required when suspicious areas require closer examina-
forming complete right heart catheterization. Special consider- tion (Figs. 22.4 and 22.5). These may be performed with balloon-
ations are made depending on the measured pulmonary artery tipped catheters following balloon occlusion. The findings
and right ventricle end-diastolic pressures. depend on the size and number of emboli, on the location of
The catheters usually used for diagnostic pulmonary the lesion—central or peripheral, and if the lesion results in
angiography range in sheath size between 5 and 7 Fr. However, partial or complete flow obstruction. Angiography may detect
lower-size 4-Fr nylon pulmonary catheter that allows flow rates intraluminal filling defects of various sizes, shapes, and loca-
of 20 mL/sec can be used to reduce access site complications. tions; abrupt arterial cutoffs and localized pruning or lack of
The pigtail catheter has multiple side holes with curled tip that branching (Fig. 22.6). The extent of these findings correlates with
[2/7/010/8:45:24] [270–274]
PULMONARY ANGIOGRAPHY 273
Figure 22.3 Normal left pulmonary artery angiogram. Figure 22.5 Selective left upper-lobe pulmonary artery angiogra-
phy performed through a balloon-tipped catheter.
Figure 22.4 Selective right upper-lobe pulmonary artery angiog- Figure 22.6 Right pulmonary artery nonselective angiography.
raphy performed through a balloon-tipped catheter. Note the presence of embolus in proximal middle- and lower-lobe
branches.
the severity of pulmonary arterial occlusion. As a correlate of COMPLICATIONS

hemodynamic compromise—cardiogenic shock—patients may Pulmonary angiography is a remarkably safe procedure in
have in addition oligemia, asymmetrical filling, prolongation of experienced hands. Reported major complications are approx-
the arterial phases, and bilateral lower-zone filling delay. imately 1.5% and include death, respiratory distress requiring
cardiopulmonary resuscitation or intubation, renal failure
requiring dialysis, or access site complications requiring trans-
LIMITATIONS fusion (18). Patients with pulmonary artery hypertension are at
Pulmonary angiography studies are challenging to interpret. higher risk of having acute right heart failure (19). However, in
The reading physician, therefore, should give some estimate of the Prospective Investigation of Pulmonary Embolism Diagno-
the degree of certainty of the diagnosis, though arbitrary, sis (PIOPED) study, increase mortality was not observed
based on a judgment of the completeness of visualization of among 755 patients who underwent pulmonary angiogra-
pulmonary circulation. phy (10). Nevertheless, due to this potential risk, pulmonary
[2/7/010/8:45:24] [270–274]
artery and right ventricular end-diastolic pressures are usually 2. Tapson VF. Acute pulmonary embolism. N Eng J Med 2008;
measured directly at the time of the procedure. If considerable 358:1037–1052.
elevation in pressure is found (right ventricular end-diastolic 3. Ginsberg JS. Management of venous thromboembolism. N Engl J
pressure above 20 mmHg), the volume and rate of contrast Med 1996; 335:1816–1828.
4. Wells PS, Hirsh J, Anderson DR, et al. Accuracy of clinical
injection can be modified or, in stable patients, the study
assessment of deep-vein thrombosis. Lancet 1995; 345:1326–1330.
deferred. Other minor complications include arrhythmias 5. Kucher N, Walpoth N, Wustmann K, et al. QR in V1—an ECG sign
related to the catheter position, right bundle branch block, associated with right ventricular strain and adverse clinical out-
nausea, vomiting, hypotension and respiratory distress related come in pulmonary embolism. Eur Heart J 2003; 24:1113–1119.
to sedation and narcotics and itching, urticaria, or other allergic 6. Brown MD, Rowe BH, Reeves MJ, et al. The accuracy of the
reactions due to contrast use (15). Critically ill and/or elderly enzyme-linked immunosorbent assay D-dimer test in the diagno-
are at higher risk of renal dysfunction (20). As opposed to prior sis of pulmonary embolism: a meta-analysis. Ann Emerg Med
large retrospective series, no cardiac tamponade was observed 2002; 40:133–144.
or myocardial perforation in the PIOPED study. This finding is 7. Dunn KL, Wolf JP, Dorfman DM, et al. Normal D-dimer levels in
likely related to the catheter choice in PIOPED study, namely emergency department patients suspected of acute pulmonary
embolism. J Am Coll Cardiol 2002; 40:1475–1478.
catheters of pigtail type instead of straight catheters.
8. Goldhaber SZ, Simons GR, Elliott CG, et al. Quantitative plasma
D-dimer levels among patients undergoing pulmonary angiogra-
phy for suspected pulmonary embolism. JAMA 1993; 270:
CONTRAINDICATIONS AND SPECIAL ISSUES 2819–2822.
There is no absolute contraindication to pulmonary angiogra- 9. Stein PD, Goldhaber SZ, Henry JW, et al. Arterial blood gas
phy. However, minimization of contrast and special care should analysis in the assessment of suspected acute pulmonary embo-
apply for high-risk patients such as those with severe pulmo- lism. Chest 1996; 109:78–81.
nary hypertension, known allergic reaction to iodine-based 10. Value of the ventilation/perfusion scan in acute pulmonary embo-
lism. Results of the prospective investigation of pulmonary
contrast media, renal insufficiency, left bundle branch block,
embolism diagnosis (PIOPED). The PIOPED Investigators.
and right heart or biventricular congestive heart failure (14).
JAMA 1990; 263:2753–2759.
Patients with history of severe allergic reactions should be 11. Rathbun SW, Raskob GE, Whitsett TL. Sensitivity and specificity
pretreated with corticosteroids and antihistamines, and use of of helical computed tomography in the diagnosis of pulmonary
nonionic low osmolar contrast agents should be considered, as embolism: a systematic review. Ann Intern Med 2000; 132:227–232.
conventional ionic monomer contrast media transiently increase 12. Schoepf UJ, Holzknecht N, Helmberger TK, et al. Subsegmental
pulmonary artery and right ventricle pressure. pulmonary emboli: improved detection with thin-collimation
multi-detector row spiral CT. Radiology 2002; 222:483–490.
13. Auger WR, Fedullo PF, Moser KM, et al. Chronic major-vessel
CONCLUSIONS thromboembolic pulmonary artery obstruction: appearance at
angiography. Radiology 1992; 182:393–398.
Although rarely needed for diagnostic purposes, pulmonary
14. Oudkerk M, Van Beek EJR, Reekers JA. Pulmonary angiography:
angiography remains the gold standard imaging modality for
technique, indications and interpretation. In: Oudkerk M, van Beek
acute and chronic PE. In addition, angiography may, on occa- EJR, Cate JWT, eds. Pulmonary Embolism. Boston, Massachusetts:
sion, be indicated in the evaluation of pulmonary hypertension, Blackwell Science, 1999:135–159.
congenital heart disease, pulmonary arteriovenous malforma- 15. Mills SR, Jackson DC, Older RA, et al. The incidence, etiologies,
tions, and pulmonary aneurysms. Finally, angiography guides and avoidance of complications of pulmonary angiography in a
all catheter-based interventions in the pulmonary circulation. large series. Radiology 1980; 136:295–299.
Pulmonary angiographic studies should be performed by 16. Hagspiel KD, Polak JF, Grassi CJ, et al. Pulmonary embolism:
experienced operators in catheterization laboratories capable comparison of cut-film and digital pulmonary angiography. Radi-
of invasive hemodynamic monitoring. Pulmonary angiography ology 1998; 207:139–145.
17. Johnson MS, Stine SB, Shah H, et al. Possible pulmonary embolus:
requires venous access, passage of the angiographic catheter
evaluation with digital subtraction versus cut-film angiography—
under fluoroscopic guidance, and contrast injection during
prospective study in 80 patients. Radiology 1998; 207:131–138.
cineangiography. In addition to technical expertise, angiographic 18. Stein PD, Athanasoulis C, Alavi A, et al. Complications and
interpretation requires great experience. validity of pulmonary angiography in acute pulmonary embolism.
Circulation 1992; 85:462–468.
19. Perlmutt LM, Braun SD, Newman GE, et al. Pulmonary arteriog-
REFERENCES raphy in the high-risk patient. Radiology 1987; 162:187–189.
1. Silverstein MD, Heit JA, Mohr DN. Trends in the incidence of deep 20. Stein PD, Gottschalk A, Saltzman HA, et al. Diagnosis of acute
vein thrombosis and pulmonary embolism: a 25-year population- pulmonary embolism in the elderly. J Am Coll Cardiol 1991;
based study. Arch Intern Med 1998; 158:585–593. 18:1452–1457.
[14/7/010/20:46:50] [275–296]
23
Transseptal catheterization
Zoltan G. Turi
INTRODUCTION currently experimental technologies will likely continue the

Left atrial access has been a challenge in cardiac catheterization trend. A list of procedures for which transseptal catheterization
since the earliest cardiac surgical procedures mandated accu- is required is provided in Table 23.1.
rate assessment of left atrial pressure in the 1940s. The occa- Transseptal technology was relatively stagnant for the
sional direct measurement was by a number of hazardous first 20 years after its introduction. In 1979, Mullins introduced
routes, including transbronchial and direct left atrial puncture. a dilator and sheath combination that provided a platform for
Modern left atrial catheterization by puncture of the interatrial advancement of a variety of catheters into the left atrium (LA),
septum is a half-century-old. It resulted from the pioneering and enhanced the safety of the procedure overall (11). The
efforts of a young medical resident at the East Orange, New primary tools for left atrial access have remained the Mullins
Jersey Veteran’s Administration Hospital, Constantine Cope, sheath, along with the Brockenbrough needle, a combination
who developed an apparatus working with the Becton with origins 30 and 50 years old, respectively. More recently,
Dickinson Company (1). This led to animal and clinical inves- however, with increasing utilization of transseptal puncture, a
tigation at Hahnemann Hospital in Philadelphia, in conjunction number of novel technologies have appeared, and will be
with Charles Bailey, the pioneering cardiac surgeon who per- discussed later in this chapter (12).
formed the first successful mitral commissurotomy in 1948 (2).
Ross and colleagues at the National Institutes of Health sub-
sequently described a similar approach (3). The technique was ANATOMY OF THE INTERATRIAL SEPTUM
inherently dangerous (4), and was used primarily for pressure The interatrial septum is embryologically derived from growth
measurement and angiography. Several factors contributed to of the septum primum and secundum toward the endocardial
its near disappearance from clinical practice. First, right heart cushions (Fig. 23.2A). During prenatal development, a circular
catheterization, developed in the 1940s, allowed indirect left opening, the foramen ovale, develops in the inferior-posterior
atrial pressure measurement through the pulmonary arterial portion of the septum secundum. As the septum primum
wedge pressure (5). Second, the introduction of balloon flota- resorbs, it continues to overlap the foramen ovale, forming a
tion catheters simplified and enhanced the safety of right heart flap that functionally acts as a valve. Because of high pulmo-
catheterization (6). Finally, the availability of echocardiography nary vascular resistance during gestation, right atrial pressure
further decreased the need for a highly invasive means of drives the valve to open and allow blood to flow from the right
assessing hemodynamics. As a result, there was a dramatic atrium (RA) to the LA. After birth, pulmonary vascular resis-
decrease in use of transseptal puncture, and by the late 1970s tance decreases, right atrial pressure falls below left atrial pres-
most adult catheterizers had either stopped doing the proce- sure, and the valve fuses to the septum secundum. Because the
dure or had never been trained to perform it. Occasional septum primum is thin and membranous compared with the
transseptal puncture was still mandated by prosthetic aortic thicker and more muscular septum secundum, the fossa ovalis
valves that were inherently dangerous to cross retrograde, in is only approximately 2 mm thick, and is recessed unless the left
particular the Bjork-Shiley tilting disk valve. However, the atrial pressure is high. It averages less than 2.5 cm2 in diameter,
declining emphasis on structural heart disease in favor of making for a relatively small target (Fig. 23.2B). A patent fora-
coronary angiography and the introduction of ever more men ovale (PFO), present in approximately one quarter of the
sophisticated cardiac ultrasound in lieu of hemodynamic stud- population, allows for easy transseptal access, but may not be
ies resulted in few operators with training, experience, or the ideal entry route to the LA as discussed subsequently.
maintenance of skills at transseptal puncture (7). Important anatomic variants include a prominent Eustachian
Although the technique continued to have adherents for valve, which can limit access to the septum from the inferior
high fidelity hemodynamic measurements, and for antegrade vena cava, and left superior vena cava (SVC), which may result
access in patients with aortic stenosis (8) or hypertrophic in a markedly enlarged adjacent coronary sinus ostium.
obstructive cardiomyopathy (9), the reappearance of transsep-
tal catheterization owed much to the introduction of percuta-
neous balloon mitral valvuloplasty in the 1980s, a procedure EQUIPMENT FOR TRANSSEPTAL PUNCTURE
that requires primarily antegrade left atrial access. The real The standard Mullins technique calls for use of the Brocken-
stimulus to the rebirth of transseptal catheterization however brough needle, a dilator and sheath as shown in Figure 23.3. The
has been in the electrophysiology laboratory, where left-sided needle is curved to allow access to the fossa ovalis; several
ablations have resulted in exponential growth (Fig. 23.1) (10). curvatures are available, with typical ones ranging from a rela-
Further utilization is related to the expansion of structural heart tively shallow 198 angle to a steeper 538 angle (Fig. 23.4A). We
disease interventions in the past decade, and a number of prefer the steeper angle for easier access to the septum in the
[14/7/010/20:46:50] [275–296]
Figure 23.1 Growth of transseptal puncture in a survey of

33 hospitals in Italy over a 12-year period. This exponential growth
was driven primarily by atrial fibrillation ablation procedures. Source:
From Ref. 10.
Table 23.1 Indications for Transseptal Puncture

I. Diagnostic studies
a. Hemodynamic assessment where noninvasive data are
equivocal
i. Aortic outflow obstruction
1. Tilting disk prosthetic aortic valve dysfunction
2. Hypertrophic obstructive cardiomyopathy
3. Inability to cross a stenotic native aortic valve
ii. Mitral stenosis—particularly prosthetic valve obstruction
b. Arterial angiography when access can only be obtained by
antegrade access through the aortic valve
II. Interventions Figure 23.2 (A) Anatomy of the interatrial septum. The fossa
a. Electrophysiology mapping and ablation ovalis is formed by the valve created in utero between the foramen
i. Atrial fibrillation ovale and the portion of the septum primum that does not resorb.
ii. Left atrial tachyarrhythmias (left atrial tachycardia/left atrial The latter functions as a flap, or valve, and allows right to left flow to
flutter)a bypass the lungs during gestation. The fusion of the valve with the
iii. Left-sided accessory pathwaysa septum secundum takes place within three months after birth, with
iv. Left ventricular arrhythmiasb patency of the foramen continuing into adulthood in approximately
b. Structural heart interventions one quarter of the population. Because the septum primum is thin
i. Balloon mitral valvuloplasty and membranous, the fossa is typically the thinnest portion of the
ii. Left atrial appendage occlusiona septal wall. (B) Anatomy of the interatrial septum, right atrial view.
iii. Percutaneous mitral valve repaira Note the proximity of the coronary sinus, immediately inferior to the
iv. Prosthetic paravalvular leak closurea fossa, and the location of the torus aorticus (the prominence at the
v. Pulmonary vein stenosis dilatation and stentinga level of the aortic cusps). Abbreviations: LA, left atrium; RA, right
vi. Percutaneous aortic valve implantationa atrium. Source: Courtesy of Dr Kalyanam Shivkumar, University of
vii. Patent foramen ovale closurea California.
viii. Left atrial pressure monitor placementa
ix. Atrial septostomy
c. Percutaneous cardiac assist
a
procedure offlabel, experimental, or pending FDA approval. atrial pressure, where the entire septum protrudes toward the
b
transseptal approach used when aortic stenosis, peripheral vas- RA, may require a less angled approach. An example of the latter
cular disease, or mechanical prosthetic valve prevents retrograde is mitral stenosis, where the fossa is both enlarged as well as
access to the left ventricle. more horizontally aligned (13). Needles are typically 71 cm in
length and taper from a proximal 18 to 21 gauge at the distal tip.
A number of needle modifications exist, including a 16-gauge
average patient. Some operators gently bend the needle (with the needle that does not taper. The Mullins sheath is typically 59 cm,
stylet in place) to provide a custom configuration, in particular to with a 67-cm dilator, and is available in a wide range of
provide a steeper angle to access the septum when the RA is diameters, typically 6 to 8 Fr for routine diagnostic procedures,
large, or the anatomy is distorted, such as when a prominent but 14 to 21 Fr for use with interventions requiring large diameter
Eustachian valve interferes with access. Patients with high left devices.
[14/7/010/20:46:50] [275–296]
TRANSSEPTAL CATHETERIZATION 277
Figure 23.3 (A) Transseptal needle, stylet, sheath, and dilator. Note the parallel orientation of the needle arrow, sheath sidearm, dilator and
needle tip. The pointer allows for orientation of the needle and the rest of the Mullins assembly as it is withdrawn from the superior vena cava
and maneuvered against the fossa ovalis. (B) The transseptal needle, dilator and sheath showing the distance markers that identify the length
of the dilator protruding from the sheath (A) and the relative distance between the position arrow of the transseptal needle and the dilator hub
when the needle tip is just inside the end of the dilator (B). The Bing stylet is seen to protrude from the needle hub at the extreme left (arrow).
The double arrow points to the site of entry of the needle into the dilator; finger compression of the space adjacent to the needle (inset) is
necessary to prevent air from being sucked into the assembly during careful drawback of blood through the needle to establish an air-free fluid
column for pressure recording and dye injection. (C) The sheath with (A) the dilator and (B) needle both fully inserted into the hub. Since there
are several commonly available lengths of the Brockenbrough needle, it is essential to confirm that the needle and sheath sizes are matched.
Source: Courtesy of Medtronic, Inc. and St. Jude Medical.
[14/7/010/20:46:50] [275–296]
Figure 23.4 A gallery of (A) transseptal needles and

(B) preformed and (C) steerable sheaths. Source:
Courtesy of St. Jude Medical.
Transseptal sheaths with a wide range of shapes are dampens the phasic excursion of the wedge derived A and V
available (Fig. 23.4B), primarily designed for electrophysiology waves. Second, a substantial phase delay is introduced. Thus,
access to various structures within or originating from the LA. as shown in Figure 23.5A, although the mean wedge and left
A transseptal sheath that allows the operator to vary the distal atrial pressures are usually identical (barring the still uncom-
curvatures (Fig. 23.4C) is widely used in ablations (14) and mon but no longer rare phenomenon of pulmonary veno-
some specialized left atrial interventions such as percutaneous occlusive disease) (16), the height of the A and V waves is
closure of prosthetic mitral paravalvular leaks (Agilis, St. Jude prone to significant underestimation when pulmonary wedge
Medical, St. Paul, Minnesota, U.S.) (15). The Agilis can be used pressures are recorded (Fig. 23.5B). Since the rate of fall in left
in the initial transseptal puncture (in which case a longer atrial pressure (negative dP/dt) is substantially diminished by
needle is required to accommodate the additional length recording across the pulmonary vascular bed, the gradient
added by the control handle) or introduced by an exchange between pulmonary wedge and left ventricular pressure is
technique after access to the LA is already achieved. artifactually higher than that noted between left atrial and left
ventricular pressure. An example is shown in Figure 23.6. This
in turn leads to substantial overestimation of the severity of
INDICATIONS mitral stenosis, especially when any degree of mitral insuffi-
Diagnostic Studies ciency is superimposed on preexisting valve obstruction. This
The initial use of transseptal puncture for direct hemodynamic phenomenon has led to overestimation of prosthetic mitral
assessment of left heart pressures has become uncommon given valve stenosis and unnecessary repeat mitral valve surgery
the small but significant risk associated with the procedure, the when wedge pressure rather than left atrial pressures has
low risk of pulmonary wedge pressure measurement, and the been relied on to assess prosthetic valve dysfunction (18).
availability of noninvasive alternatives in most cases. Never-
theless, in settings where echocardiographic data are equivocal, Aortic Outflow Obstruction
transseptal catheterization remains an important alternative Technical failure to cross the aortic valve retrograde should be
means of access to the LA and the left ventricle (LV). rare and is usually associated with severe calcification and an
eccentric orifice; this can be addressed by the alternative of
Mitral Valve Hemodynamics: Wedge Vs. Left Atrial Pressure transseptal antegrade access (8). In addition, some mechanical
Pulmonary artery wedge pressure differs from left atrial pres- valves should not be crossed retrograde because of the risk of
sure measured directly in several clinically important ways. catheter entrapment and potential difficulties with extracting
First, the additional resistance in the pulmonary vascular bed the catheter (19). Furthermore, entrapment of the catheter in a
[14/7/010/20:46:50] [275–296]
Figure 23.5 (A) Left atrial and simultaneous pulmonary

artery wedge pressures, 40-mm Hg scale. Note that the
height of the A wave in this patient is higher on the left
atrial pressure tracing, while the V wave is similar. An
arrow points to diastasis between the pressure waveforms;
the mean pressures are typically identical, with the excep-
tion of patients with pulmonary venous outflow obstruction.
(B) Simultaneous left atrial and pulmonary artery wedge
pressures at 40-mm Hg scale in a patient with severe
prosthetic mitral perivalvular regurgitation. Although the
mean pressures are the same, note that the peak left atrial
pressure (diamond-shaped arrow) is 42-mm Hg, while the
peak wedge pressure is 24-mm Hg, with a significant
phase delay. The gradient between wedge pressure (dou-
ble arrows) and left atrial pressure (single arrows) is sub-
stantial, but entirely artifactual, induced by measuring the
wedge pressure indirectly across a high resistance system:
the pulmonary capillary bed. Abbreviations: LA, left atrium;
PCW, pulmonary capillary wedge.
mechanical aortic valve will distort hemodynamics since a Furthermore, the frequency of these initial findings has not
leaflet is pinned in an open position resulting in artifactual been confirmed (24). Importantly, left ventricular and central
aortic insufficiency. Simultaneous measurement of left ventric- aortic pressures can be measured simultaneously using several
ular pressure via antegrade access to the LV and retrograde techniques that avoid transseptal puncture, including use of a
access to the central aorta does address two concerns: artifact catheter designed for simultaneous measurement using a ret-
induced by measurement of left ventricular pressure against rograde technique (Langston, VascularSolutions, Minneapolis,
femoral artery sheath sidearm pressure and the potential for a Minnesota, U.S.) (25), a simultaneous catheter and pressure
small reduction in systemic pressure created by the obstruction wire method (26), and dual arterial punctures. Overall, routine
of the aortic valve by the catheter itself (Carabello’s sign) (20); transseptal puncture for hemodynamic evaluation of aortic
the latter is uncommon with 6-Fr catheters. Using femoral stenosis does not appear warranted.
artery rather than central aortic pressure does result in under- A second setting for use of transseptal puncture to assess
estimation of the transvalvular gradient regardless of whether left ventricular hemodynamics via an antegrade approach has
the left ventricular pressure is measured via transseptal been in patients with hypertrophic cardiomyopathy with
access (21) or by retrograde left ventricular entry (Fig. 23.7) obstruction. The transseptal route can avoid catheter entrap-
(22). There is also a concern that retrograde passage of catheters ment and artifact (27); pullback to the mitral valve allows
across the aortic valve is associated with cerebral emboli (23) in differentiation between apical or midcavitary obliteration and
over 20% of patients, although most are not clinically apparent. true outflow obstruction (Fig. 23.8).
[14/7/010/20:46:50] [275–296]
Figure 23.6 The patient whose hemodynamics are presented in this figure was initially misdiagnosed as having severe mitral stenosis and
referred for balloon mitral valvuloplasty. The tracing at left demonstrates a substantial gradient measured using a pulmonary arterial wedge
pressure against left ventricular diastolic pressure (40-mm Hg scale). Two clues to the true nature of the actual diagnosis are present: first,
note the V wave peak is more than twice the mean wedge pressure, and second, note the presence of diastasis at end-diastole. In contrast, a
simultaneous left atrial and left ventricular pressure measured in the same patient clarifies the hemodynamics: the gradient is primarily due to
the sizeable V wave, with decompression of left atrial pressure only slightly delayed by the presence of mild mitral stenosis; diastasis can now
be seen to occur much earlier in diastole. Source: From Ref. 17.
Transseptal Access to the Arterial Circulation tion from catheter manipulations in the LV and aorta. In addi-
In selected patients, access to the arterial circulation cannot be tion, some patients are not suitable for retrograde ablations,
obtained by any practical means. These patients typically have such as those with severe peripheral vascular disease and those
obstructions of the subclavian arteries and aorta, and limited with mechanical aortic valves. Treatment of atrial fibrillation
approaches such as carotid or lumbar transaortic access are and other atrial arrhythmias by catheter ablation in the LA and
almost always high risk. An alternative has been transseptal the pulmonary vein ostia is now widely performed (32). Dual
access, typically for aortography. Coronary angiography is tech- transseptal access to the LA is required for some procedures,
nically difficult, but has been described (28). In addition, carotid and is achieved either by performing two separate transseptal
stenting has been performed via the transseptal route in Takayasu’s punctures or by placing two catheters across a single puncture
arteritis (29) as well as in the setting of a Type III aortic arch, site; the safety implications are discussed subsequently.
where neither femoral nor arm access was deemed suitable
(30). Structural Heart Interventions
Since the introduction of percutaneous balloon mitral valvulo-
plasty with a device originally designed for atrial septos-
Interventions tomy (33), transseptal catheterization has been utilized for a
Electrophysiology variety of structural heart interventions. In patients with rheu-
Ablation for atrial fibrillation and other left-sided electrophy- matic mitral stenosis with favorable anatomy, balloon dilatation
siologic interventions has accounted for the largest segment of is the procedure of choice (34). Several innovative technologies
the growth of transseptal puncture in the past two decades require left atrial access, including percutaneous mitral valve
(Fig. 23.1). Treatment of preexcitation syndromes transitioned repair (35), left atrial appendage occlusion (36), implantation of a
from medical therapy to an investigational retrograde left left atrial pressure monitoring device (37), and closure of pros-
ventricular approach to a transseptal access based procedure thetic paravalvular leaks (15). PFO closure, in the setting of a
as first line therapy (31). The advantages of the transseptal long foramen tunnel, is sometimes performed by placing a
approach include avoidance of several potential complications closure device through a transseptal puncture placed across
associated with retrograde left atrial access. These include large rather than through the tunnel (38). A consequence of ablation
femoral artery puncture related adverse events, potential dam- procedures, scarring of the pulmonary veins, can result in pul-
age to the aortic valve or submitral apparatus from extensive monary venous obstruction that can be treated by balloon dila-
catheter manipulation, and perforation, dissection or emboliza- tation and stenting (16), also by transseptal approach.
[Shaji][D:/informa_Publishing/Mukherjee_2400046/z_production/z_3B2_3D_files/978-1-8418-4664-
4_CH0023_O.3d] [16/7/010/16:34:8] [275–296]
Figure 23.7 (A) Severe aortic valve stenosis with left ventricular, central aortic and left atrial pressure on 200-mm Hg scale. The left
ventricular pressure was obtained by antegrade introduction of a pigtail catheter into the LV through a Mullins sheath. Aortic pressure was
measured through a pigtail catheter in the central aorta (Ao). The peak-to-peak gradient is approximately 60-mm Hg peak to peak. The arterial
pressure upslope (dP/dt) is markedly blunted. The arrow points to the start of the upstroke of central aortic pressure coincident with opening of
the aortic valve. (B) Left ventricular, femoral artery and left atrial pressure on 200-mm Hg scale in the same patient obtained moments after
the tracing in 23.7A. The gradient is underestimated and the arterial pressure upslope (dP/dt) appears relatively well preserved, both the
result of recruitment of harmonics as the pulse waveform travels distally through the arterial tree. The arrow points to the substantial delay in
pressure upstroke seen when femoral artery rather than central aortic pressure is compared with the left ventricular pressure. Although left
ventricular-femoral artery gradient measurement is the most commonly used, it leads to inaccurate estimation of aortic valve area. Several
less invasive means of obtaining the gradient are discussed in the text. Abbreviations: LA, left atrium; LV, left ventricle; FA, femoral artery.
Percutaneous balloon valvuloplasty of the aortic valve via the Inoue balloon is typically not long enough for retrograde access
antegrade approach has a number of adherents, in particular to the aortic valve but is of sufficient length to place trans-
because of superior immediate hemodynamic results with the septally. However, long-term benefit of this technique compared
Inoue balloon compared with cylindrical balloons (39). The with the conventional retrograde approach has not been shown
[14/7/010/20:46:50] [275–296]
Figure 23.8 The benefit of transseptal access for invasive

assessment of left ventricular outflow track obstruction. In
the figure at the left, the gradient is caused by catheter
entrapment in a hypertrophic apex with cavitary obliteration
in systole. As the catheter is pulled back from the apex
toward the outflow track, the gradient disappears. In the
tracing at right, the outflow track narrows, and the gradient
persists until the catheter is pulled back into the left atrium.
Aorta is marked as “Ao.” Abbreviations: LA, left atrium; LV,
left ventricle. Source: From Ref. 27.
(40). Finally, percutaneous aortic valve implantation was initially more complex include a thickened or fibrotic septum and atrial
performed antegrade via the transseptal route (41). This septal aneurysm (which increases risk of perforating the left
approach has been abandoned because of complexity and asso- atrial free wall). Repeat transseptal puncture has been reported
ciated morbidity, in particular trauma to the submitral apparatus as more difficult, in part because of increased atrial septal
(42) caused by shortening of the catheter loop in the LV that thickness (45). Finally, the presence of a prior atrial septal
effectively “filleted” the chordae and papillary muscles. defect (ASD) closure either by surgery or percutaneously
placed device has been addressed using intracardiac (ICE) or
transesophageal (TEE) echocardiographic guidance. Limited
Cardiac Assist experience in these setting has suggested that transseptal punc-
Transseptal access is required for a percutaneous cardiac assist ture can be performed with reasonably high success and low
device that provides extracorporeal circulation by shunting oxy- complication rates (46).
genated blood from the LA using a 21-Fr transseptal cannula that
needs stable positioning in the LA (TandemHeart, CardiacAssist,
Inc., Pittsburgh, Pennsylvania, U.S.) (43). Blood is then pumped TECHNIQUE
into a large femoral arterial cannula. The technology provides Preprocedure Evaluation
circulatory support during high-risk percutaneous interventions, Knowledge of the anatomy of the septum, the LA and its
perioperatively during cardiac surgery, and temporarily aug- appendage, and the RA can be extremely helpful prior to
ments cardiac output in a variety of settings (Table 23.1). transseptal puncture. This is particularly true in patients pre-
disposed to thrombus. Although most clot occurs in the left
atrial appendage, it occurs with much higher frequency in the
CONTRAINDICATIONS rest of the LA in the setting of atrial fibrillation and rheumatic
Relative contraindications to transseptal puncture include heart disease (47). Thrombus along the septum, on prosthetic
uncorrected anticoagulation as well as significant coagulopathy valves, and on pacemaker wires adds substantial risk, and most
(although some operators puncture with the patient fully anti- operators consider these absolute contraindications to trans-
coagulated as discussed subsequently) (44), thrombus in either septal puncture. However, safe performance of procedures
the right or LA, and anatomic abnormalities that alter land- such as balloon mitral valvuloplasty despite the presence of
marks and interfere with septal access. The latter include severe appendage thrombus has been described (48). The relative
kyphoscoliosis, giant LA, prominent Eustachian valve, left thickness of the septum, aneurysmal excursion, deviation into
SVC, vena caval interruption, and an azygous continuation of the RA (more common with left atrial hypertension such as is
the inferior vena cava. While transseptal puncture has been seen with mitral valve stenosis) (49), and presence of a PFO are
described in most of these settings, the risks are augmented. all potentially important variables to know prior to attempting
Some other features that make transseptal puncture technically the transseptal.
[14/7/010/20:46:50] [275–296]
Figure 23.9 (See color insert) (A) Three-dimensional echocardiography to demonstrate apposition of the transseptal apparatus with tenting
of the septum and (B) the appearance after needle and sheath entry into the LA. Intracardiac or transesophageal echo is used for echo
guidance by most operators. Abbreviations: CS, coronary sinus; FO, foramen ovale; LA, left atrium; RA, right atrium. Source: From Ref. 55 by
permission of the Oxford University Press and the European Heart Rhythm Association.
Imaging remains the right femoral route. Left femoral venous access
The most common techniques for adjunctive imaging are requires negotiating the additional curvature of the left iliac
ICE (49) and TEE (50), both of which can blunt the learning vein, in particular as it enters the inferior vena cava, and
curve for transseptal puncture, and provide for a margin of provides technically more difficult access to the interatrial
safety even for highly experienced operators. Transthoracic septum; it is nevertheless quite feasible. We place a short 7-Fr
echo alone has been used as an adjunct when transseptal sheath in the right femoral vein, and perform right heart
puncture with fluoroscopic guidance alone was difficult (51), catheterization if clinically warranted. This helps identify
but this technique has largely been supplanted. TEE preproce- hemodynamics before transseptal puncture, including the pres-
dure in patients at high risk of left atrial thrombus is a Class I sure to expect when the needle enters the LA, and establishes
indication (34), and both TEE and CT angiography have been the baseline hemodynamics should a subsequent suspicion of
used to assess for thrombus (52,53). The consensus statement tamponade occur. We also fluoroscope the left heart border in
published by the Heart Rhythm Society lists TEE as the “gold the anterior-posterior view, to note the degree of motion seen
standard,” with a recommendation that all patients with atrial before puncture. At this point, we advance a pigtail catheter
fibrillation at the time of ablation should be screened with a retrograde to the aortic valve, usually from the femoral artery,
TEE (54). More recently, real-time three-dimensional echo guid- although a radial approach can also be used. Because the fossa
ance has been utilized to facilitate transseptal puncture is usually below the level of the aortic cusps, placement of the
(Fig. 23.9) (55). Preprocedure CT angiography, in conjunction pigtail at the aortic valve is an important adjunct to prevent
with electroanatomic mapping (56) has been used as an adjunct entry into the aorta or perforation of the RA secondary to a high
to transseptal puncture, and an animal model using intrapro- puncture (Fig. 23.10) (61). Electrophysiologists performing
cedural magnetic resonance imaging to guide laser driven transseptal puncture will frequently place a catheter in the
transseptal puncture has been described as well (57). coronary sinus to provide an alternative or additional anatomic
In certain settings, such as pregnant patients where the landmark (Fig. 23.2B) (62). The pigtail catheter is connected to a
operator has deemed fluoroscopy to be hazardous, the proce- pressure manifold, and systemic pressure is continuously dis-
dure has been done with echo guidance alone (58) or with played during the procedure, from pretransseptal puncture to
fluoroscopy plus echo designed to minimize radiation (59). at least five minutes after withdrawal of the transseptal sheath,
Since minimal radiation is required for a transseptal, most to ensure that sheath withdrawal does not unmask a stitch
operators continue to use fluoroscopy in pregnant patients to perforation (discussed subsequently) or other errant puncture
minimize procedure duration and for the added margin of that may result in early tamponade.
safety provided by the additional imaging guidance (60).
Maneuvering to the Septum

TRANSSEPTAL PROCEDURE After withdrawal of the right heart catheter, the venous sheath
Preparation is carefully flushed. The Mullins assembly is potentially quite
Although transseptal puncture has been performed by venous thrombogenic and should not be introduced into the vein until
access from a variety of approaches, including the internal all the equipment is completely prepared and the operator is
jugular, subclavian and even hepatic veins, the standard ready to perform the puncture. A 0.032-in J-tipped guidewire is
[14/7/010/20:46:50] [275–296]
Figure 23.10 (A) The Mullins assembly has been advanced into the SVC (dashed arrow), a pigtail catheter (white arrow) is in the aortic
cusp, and a pulmonary artery catheter is in the left pulmonary artery (the pulmonary artery catheter was placed from the left femoral vein).
(B) The Mullins assembly has been withdrawn from the SVC into the right atrium. The tip (white arrow) is above the bottom of the pigtail
catheter (broken white line), and is too high to attempt puncture. If the needle is advanced in this position, there is an increased chance of
inadvertent entry into the aorta. Both images are in anterior-posterior view. Abbreviation: SVC, superior vena cava.
then advanced through the femoral venous sheath (some proximal, there is a risk of the dilator kinking when the
Mullins assemblies will accept a 0.035-in wire); when advanced assembly is pressed against the septum and can result in
from the femoral vein directly (rather than inside a catheter), perforation of the dilator when the needle is advanced. If it is
the wire tends to enter the SVC easily. If introduced through too distal, there is a risk of inadvertent needle deployment and
the Mullins dilator, the curvature tends to steer the wire away perforation of the SVC or RA. The operator should maintain his
from the ostium of the SVC, and more manipulation is typically fingers in the space between the needle arrow and the dilator
required. Once the wire is in place in the SVC, the venous hub to prevent inadvertent movement of the needle proximally
sheath is removed and the Mullins dilator and sheath are or distally. The alignment of the Mullins sheath, dilator and
advanced; some operators place the assembly into the left needle should be parallel (Fig. 23.3A). Care should be taken to
subclavian vein. The dilator is flushed, taking care to make maintain this alignment throughout the subsequent movement
sure there is venous return, since aggressive flushing of of the Mullins assembly.
the dilator buried in the vessel wall can cause local dissection. Once properly flushed and an undamped pressure trac-
The Brockenbrough needle and Bing stylet (Fig. 23.3A) are ing is seen on the monitor, the entire assembly is pulled down
advanced through the dilator, with care taken to allow into the RA with the needle pointer aimed in a posterior medial
the needle to rotate freely as it traverses toward the SVC. If direction (approximately 4–5 o’clock, with 6 o’clock being
the needle is held rigidly during advancement and prevented straight posterior). The tip of the dilator can usually be seen
from rotating freely, there is a small risk of perforation; in to deflect off the SVC-right atrial junction, and then to fall
addition, it places pressure against structures that abut the vena below the limbus. We typically pull it back an additional
cava, particularly Glissen’s capsule of the liver, and can cause centimeter and advance into the fossa. At this point, unless
considerable discomfort. The Bing stylet helps prevent perfo- the fossa is displaced toward the RA by elevated left atrial
ration of the sheath and dilator as the needle is advanced. Once pressure, resistance should be felt.
the needle and stylet enter the RA, the needle can be turned to
face toward the patient’s left (3 o’clock) as it is advanced up
into the SVC, and the stylet withdrawn when the needle is a Localization at the Fossa Ovalis
few centimeters from the tip of the dilator. The needle is then Several maneuvers can confirm appropriate location of the
connected to extension tubing attached to the manifold. Flush- dilator prior to attempted puncture. Fluoroscopy should dem-
ing the needle requires some care while drawing back on the onstrate that the tip of the Mullins assembly is below the pigtail
column of fluid in the needle and should be done slowly with that was placed in the aortic root; the relevant relationship of
the operator using his fingers pressed against the point where the dilator to the pigtail catheter and anatomic structures can be
the needle enters the dilator hub to prevent air from entering seen in different views in Figures 23.10 and 23.11. Although
around the needle (Fig. 23.3B). The needle should be kept at puncture can also be performed in the 208 RAO view (Fig. 23.12)
least a few millimeters proximal to the dilator tip. If it is too (63) or the anterior-posterior view as well, in our experience the
[14/7/010/20:46:50] [275–296]
Figure 23.11 The correct position for transseptal puncture in the (A) anterior-posterior and (B) 208 right anterior oblique planes. The fossa
ovalis has been stained (white arrow), and the Mullins assembly can be seen to have entered the fossa. Note that the target for transseptal
puncture is approximately 1 to 2 cm below the bottom of the aortic cusp (as demarcated by the location of the pigtail catheter in the aorta). In
the 208 right anterior oblique view, the fossa is typically located below the center of a line drawn from the right anterior free wall to the bottom
of the aortic cusp.
Figure 23.12 The Mullins assembly in simultaneous (A) anterior-posterior and (B) 908 left lateral views; the needle is too low and too
posterior to consider attempted transseptal puncture. In the image on the left, the tip of the catheter can be seen to be well below the aortic
cusps; in the image on the right the catheter tip abuts the posterior free wall of the right atrium. Needle advancement in this position would
lead to pericardial entry and risk of tamponade. This patient had severe mitral stenosis and was undergoing transseptal puncture for mitral
valvuloplasty; note the double density caused by the enlarged left atrium (short arrows).
best guidance to avoiding the left atrial free wall is the 908 lateral with great care afterward, since brachial plexus injury can result
(Fig. 23.13). (Adequate imaging in this view requires that the if inadvertent traction is placed on the arms.) The relationship of
patient’s arms be raised out of the field; if the procedure is done the fossa ovalis to the aortic cusps and posterior LA is most likely
under anesthesia the arms should be moved prior to induction or to be fixed in the center of a line drawn at a 458 angle between the
[14/7/010/20:46:50] [275–296]
Figure 23.13 The correct position for transseptal puncture in the left lateral view. (A) The tip of the transseptal assembly (double arrow) is
tenting the stained fossa which is located halfway along a dotted line drawn at a 458 angle from the pigtail in the aortic root (solid white arrow)
to the free wall of the LA (dashed arrow). The fossa is tented. (B) The needle has been advanced into the LA. Left atrial entry was confirmed
by pressure measurement and dye injection, and the tip of the dilator can be seen to have advanced through the fossa as well. Abbreviation:
LA, left atrium.
aortic valve and the left atrial posterior wall. If the catheter is low,
it can be moved into position with a gentle to-and-fro “wind-
shield wiper” maneuver: clockwise turns the assembly posteri-
orly and counterclockwise anteriorly. Aggressive manipulation
should be avoided; besides trauma to the RA, patients in sinus
rhythm predisposed to atrial arrhythmias, such as those with
mitral stenosis and dilated LA, are prone to develop atrial fibril-
lation. With normal anatomy, the dilator tip will be tented against
the fossa; contrast injection to “tag” the septum is benign and can
assist visualization but is not ordinarily necessary (Figs. 23.11 and
23.13). If the operator is uncertain of location, however, this is a
useful maneuver.
If ICE (49) or TEE is performed as an adjunct to the
procedure, tenting of the fossa (Fig. 23.9) should be seen. The
advantages of intraprocedural echocardiographic guidance
have been extensively demonstrated (64), although operators
should be sufficiently versed in transseptal anatomy that they
are not unduly dependent on echo guidance. However, addi-
tional benefits of ICE or TEE beyond anatomic guidance for
transseptal puncture include early warning of pericardial effu-
Figure 23.14 One scheme for transseptal puncture in the 408 right
sion. A review of tamponades during ablation procedures
anterior oblique view. The fossa is typically located 1 to 3 cm below
revealed echocardiographically apparent pericardial effusion
the horizontal line drawn from the wall of the aorta to the posterior
before hemodynamic compromise in 11 of 13 patients (65). A free wall. A number of different schemes to identify the location of
characteristic electrogram recorded through the Brockenbrough the fossa have been published. Source: From Ref. 67.
needle and associated with the fossa ovalis has been described
(66), although this technique is rarely if ever used. A number of
other schemes (Fig. 23.14) for assessing location of the septum
and the fossa have been described and include right atrial
angiography in a variety of views, including visualization of it is possible to advance into the LA through the foramen ovale
the LA in the levo phase of the injection. in a significant percent of patients, variably described as 25%
At this point, with pressure applied by the dilator tip to 90% (68). Although a classic PFO may not be present,
against the fossa ovalis, and without deployment of the needle, prolonged gentle pressure may allow “peeling” apart the
[14/7/010/20:46:50] [275–296]
overlapping layers of septum secundum and residual septum combined entry of the dilator and needle into the LA, a tech-
primum when the foramen does not ordinarily open with the nique best avoided because needle entry alone is usually
usual maneuvers. Catheter passage through a PFO is suitable benign, while errant entry of both needle and dilator is much
for diagnostic catheterization but causes problems for many more likely to lead to tamponade.
interventional procedures, since the foramen tunnel restricts The operator has several means of confirming left atrial
pivoting of the sheath, and the orientation is relatively superior entry. A tactile sensation of the catheter popping through the
and posterior. In some cases, passage without needle deploy- septum is usually felt. Entry of the needle into the LA should be
ment is not through the foramen tunnel but directly across an immediately apparent when a distinctive left atrial pressure is
extremely thin fossa, the latter in keeping with the membra- seen on the monitor (Fig. 23.15). When the fossa is thickened, or
nous nature of the septum primum that forms the embryonic the operator advances into the thick septal wall outside of the
fossa valve. fossa, a damped pressure is usually seen; in older patients
Although transseptal puncture at the center of the fossa is the septal wall can be fibrosed or calcified, in which case the
usually ideal, some interventions require high or low puncture. waveform may occasionally appear to be a reflection of a
The former is desirable when a relatively perpendicular plane slightly damped left ventricular pressure. In some cases, a
of access is required toward the mitral valve, such as for higher velocity of needle entry will allow for penetration of
percutaneous mitral valve repair with a clip device (35), or the thickened or fibrosed septum; it is occasionally necessary to
left atrial appendage occlusion (36). Low puncture is desirable pass the needle and dilator together in this setting. It is essential
when a relatively shallow entry to the mitral valve is ideal, that the operator be reasonably certain of the needle’s location
such as was the case for percutaneous metallic mitral against the fossa before committing to such combined entry. In
commissurotomy (69). addition to tactile sensation and pressure monitoring, contrast
injection will confirm needle location in the LA. Oxygen satu-
ration can also be obtained through the transseptal needle. If
Puncture and Device Advancement ICE or TEE is being performed, the needle and catheter are
Once the operator has ascertained that the location is suitable, it usually visualized in the LA (Fig. 23.9), the tenting of the
is essential that only the needle and not the rest of the assembly septum resolves, and saline or contrast injection appears as
perforate the septum (Fig. 23.13B). A video demonstrating “bubbles” in the LA on echo. Some operators use a coronary
transseptal entry using a 908 lateral view can be downloaded guidewire immediately after needle puncture to confirm entry
from an article by Cheng and colleagues (70). The technique into the LA (Fig. 23.16) (71).
shown does not include use of the pigtail catheter in the aortic At this point we rotate the image intensifier to visualize
root to identify the level of the aortic cusps, and features catheter advancement in an anterior-posterior plane. If the
Figure 23.15 One of several modalities for confirming catheter entry into the LA. Damped right atrial pressure (RA, double arrow) was
recorded while the dilator tip was against the fossa. Immediately after transseptal puncture, the phasic waveform characteristic of left atrial
pressure is seen (LA, black arrow). Right and left atrial pressures, 40-mm Hg scale; aortic pressure (AO), 200-mm Hg scale. Abbreviations:
LA, left atrium; RA, right atrium.
[14/7/010/20:46:50] [275–296]
Figure 23.16 Transseptal assembly advanced into the left atrium

over a 0.014-in coronary angioplasty guidewire (white arrow). The Figure 23.17 The transseptal needle (dashed arrow) is seen to
needle was advanced only to a point just across the septum, and left approach the roof of the left atrium. Contrast pools (double arrow)
in place to support advancement of the dilator and sheath. This along the curved wall of the left atrial roof (small arrows). Even a few
patient has prosthetic bi-leaflet aortic (A) and mitral (M) valves, and millimeters of further advancement would have resulted in perfora-
a patent ductus arteriosus plug (double arrow) adjacent to the mitral tion and possible tamponade or hemothorax.
annulus used to treat a prosthetic paravalvular leak. A transeso-
phageal echocardiographic probe and endotracheal tube can also
be seen in the image. The coronary guidewire was introduced
through the Brockenbrough needle, and provides visual clues as
to distance of the dilator tip from adjacent cardiac structures. The
image was recorded at the beginning of a procedure where a
second paravalvular leak was closed.
or dilator (with no guidewire of any kind used), and an abrupt
forward movement frequently occurs as the sheath penetrates
the septum over the dilator, especially if the sheath is over-
coming resistance caused by a thickened septal wall. For certain
procedures, such as percutaneous balloon mitral valvuloplasty,
septum is thin, it may be possible to advance the dilator and sheath entry is optional, since the Inoue guidewire can be
sheath without further advancement of the needle. If the advanced through the dilator alone.
septum is thick, it may be necessary to advance the entire
assembly to have enough support to prevent the sheath from Anticoagulation
buckling as it advances against the septum. Care needs to be There has been a dramatic change in the approach to anti-
taken at this point since it is possible to have the entire assem- coagulation for transseptal puncture in the last several years.
bly prolapse back into the RA if the septum offers resistance. Of This represents a dichotomy between interventional cardiolo-
greater concern is the risk of perforating the LA as the assembly gists performing structural heart disease interventions who
is advanced. Continuous pressure monitoring through the have been concerned predominantly about tamponade and
needle with immediate discontinuation of advancement if pres- electrophysiologists performing ablations whose primary focus
sure damping is noted, as well as small puffs of contrast to has been the avoidance of thrombus formation and stroke.
identify if the assembly is near to the left atrial wall will protect Ablations tend to be of longer duration and have several features
the LA (Fig. 23.17). If the septum is relatively soft, it may be that increase the predisposition to clotting, including platelet
possible to withdraw the needle completely once the dilator has activation by the catheters themselves and char formation that
crossed the septum, in which case a J-tipped guidewire can be triggers thrombus formation. Some operators, including many
placed through the dilator to allow safe advancement of the electrophysiologists, give at least a small amount of heparin
sheath and dilator assembly. Alternatively, if support from the before the puncture (73) on the basis of data suggesting a lower
shaft of the needle is required, a 0.014@ coronary guidewire can rate of intraprocedural left atrial thrombus, while others perform
be advanced through the needle into the LA or a pulmonary transseptal puncture with patients fully heparinized (70) or
vein, to visualize approach to the left atrial wall, and provide maintained on therapeutic oral anticoagulation (44). A compar-
some protection from perforation (Fig. 23.16) (72). A significant ison of warfarin discontinuation plus bridging of anticoagulation
risk for perforation occurs during blind advancement of needle with enoxaprin versus maintenance of therapeutic range oral
[14/7/010/20:46:50] [275–296]
anticoagulation during the periprocedure period did not demon- of diluted contrast in the sheath so that as a device is intro-
strate a difference in complication rates (74). duced, bubbles in the column are readily detected under fluo-
In contrast, interventional cardiologists have focused on roscopy, in which case the device can be withdrawn and the
avoidance of tamponade, with most deferring transseptal punc- sheath reprepared as necessary. The Mullins sheath needs to be
ture if the international normalized ratio is greater than 1.6 or flushed regularly, or a drip maintained with a system designed
patients are heparinized. However, stroke secondary to throm- to prevent air from entering the column of fluid.
bus formation is an important source of morbidity and mortal-
ity during prolonged structural heart disease interventions as
well, an example being closure of prosthetic paravalvular
leaks (75). And while aggressive anticoagulation may help COMPLICATIONS
prevent thrombus formation, it is important to note that the Complications of transseptal puncture are significant and
most common single cause of peritransseptal mortality with potentially life threatening (Table 23.2). The event rates are
structural heart disease interventions as well as ablation proce- quite variable, and are influenced by several important factors,
dures remains pericardial tamponade (76). Importantly, even including level of anticoagulation, duration of the procedure,
when anticoagulation is deferred until after puncture, the size of catheter used, intracavitary pressure, status of pericar-
tamponade rate is higher than 1% (77). The exact peri- and dium (intact or removed), use of echocardiographic guidance,
postprocedure anticoagulation management of these patients and most importantly the operator learning curve (83). The
remains a focus of investigation (78). major complication rate is far lower in diagnostic (1%) (84) than
If patients are not anticoagulated at the time of needle interventional procedures (4%) (85), likely because of less
entry into the LA, transseptal puncture should be performed aggressive anticoagulation in most diagnostic studies, shorter
expeditiously to allow anticoagulation as soon as the LA has duration of diagnostic procedures, and the addition of morbid-
been entered to prevent thrombus formation on the catheter tip ity associated with the interventions themselves. In the era
or guidewire. The target for activated clotting time (ACT) has before ablation procedures were common, the incidence of
increased in the recent literature to a range of 300 to 400 seconds tamponade was more than 10 fold higher (1.2%) than the
(79); comparison of two groups with ACTs of 250 to 300 seconds incidence of stroke (0.1%) (84). While the profile of complica-
versus ACT greater than 300 seconds demonstrated a reduction tions seen after transseptal puncture for structural heart disease
of catheter associated thrombus from 11% to 3% (80). The con- interventions may be significantly different than that seen with
sensus statement by the Heart Rhythm Society recommends electrophysiology procedures, as the complexity and duration
continuous heparin infusion at 10 units/kg/hr with ACT of the former increase, these differences may be muted.
checked every 10 to 15 minutes until therapeutic anticoagulation
is achieved, and then every 30 minutes thereafter (53). Despite the
relatively aggressive anticoagulation regimen, stroke continues to Tamponade
be a factor after transseptal puncture; in the ablation literature, As discussed, the most dreaded complications of transseptal
transient ischemic events and stroke continue to occur in the 0.5% puncture remain pericardial tamponade and clot or air embo-
range (77). However, transcranial Doppler does demonstrate that lization. The incidence of tamponade is highly variable, but
a significant portion of microembolic signals are unrelated to the remains a significant concern in a wide variety of procedures
puncture itself (81). associated with transseptal puncture (86), and has been
described to range from 0.5% to 4% in patients undergoing
percutaneous balloon mitral valvuloplasty (87). In patients
Sheath Management undergoing ablation, tamponade has been described as occurring
Handling of the transseptal sheath for the rest of the procedure in up to 6% of cases (54), associated with the type (linear) and
needs to follow certain basic principles. Wires and catheters
should be withdrawn from the Mullins sheath slowly, since
abrupt negative pressure caused by rapid catheter or wire
withdrawal can induce air to enter the sheath. Similarly, flush-
Table 23.2 Complications of Transseptal Puncture
ing should be performed with care to avoid sucking air in
through the sheath diaphragm; we prefer to allow back bleed- I. Mechanical
ing through the sheath sidearm and do not put negative pres- a. Perforation
sure on the sheath since the diaphragm can leak air into the i. Pericardial tamponade
fluid column when negative pressure is applied. Passive air ii. Hemothorax
embolization due to the gradient between atmospheric pressure b. Persistent atrial septal defect
c. Right to left shunting
and low left atrial pressure, a phenomenon most common in
d. Aortoright atrial fistula
the setting of sedation and exacerbated by placement of a
II. Embolization—peripheral or central nervous system
guidewire through a sheath hemostatic valve, is an important a. Thrombus
and relatively common occurrence, and requires careful tech- b. Air
nique to avoid (82). If the left atrial pressure is low, it may be III. Arrhythmias
necessary to suspend the end of the Mullins sheath below the a. Atrial fibrillation
patient to ensure that pressure at the open sidearm is lower b. Supraventricular tachycardia
than left atrial pressure. Some operators keep the proximal end c. Inferior ST segment elevation
of the sheath inside a large bowl of saline or other solution. d. Bezold-Jarisch reflex
Introduction of air during introduction of devices through the e. Sinus node dysfunction
IV. Death
Mullins sheath is a common occurrence. We maintain a column
[14/7/010/20:46:50] [275–296]
Air as well as clot embolization remain important sources of

morbidity (91); setting up a continuous flush through the
sheath during extended procedures has been described as
preventing air and clot embolization, and the stroke risk
appears to be lower when a high-flow state is maintained
through the transseptal sheath (92). If bolus flushing of the
sheath is performed, consideration should be given to using a
high heparin concentration, since there is evidence that throm-
bus formation is inversely related to the concentration (93).
Stroke and transient ischemic attack secondary to transseptal
catheterization have been widely reported to occur in the range
of approximately 1% with values as high as 5% in patients with
ablation procedures for atrial fibrillation (94). A phenomenon
that appears to be migraine has been described in approxi-
mately 0.5% of patients, and may occur up to one week after
transseptal puncture (95).
Persistent Atrial Septal Communication

Figure 23.18 Mechanism of a stitch perforation. The transseptal The creation of a communication between left and right atria
puncture enters the LA passing outside the septal wall into extrac- commonly results in postprocedure shunting (96). Use of
ardiac space, typically adipose tissue. The Mullins assembly and
balloon dilatation to allow larger sheaths to enter the septum
any catheters subsequently placed across the fenestration may
prevent tamponade until the catheters are removed at the end of
increases the risk of persistent ASD (97) as well as tamponade.
the case. Some operators leave a guidewire across the septum after Some transseptal flow after the procedure is almost always
sheath withdrawal to confirm absence of fluid accumulation for a few seen on TEE, and correlates with size of the puncture as well
minutes prior to sheath withdrawal. Abbreviations: LA, left atrium; as driving pressure determined by disparity between left and
RA, right atrium. right atrial pressures. Thus, early in the era of percutaneous
balloon mitral valvuloplasty, when the septum was routinely
dilated with peripheral angioplasty balloons ranging in size
from 5 to 10 mm, and two transseptal sheaths were usually
placed, left-to-right flow was detected by TEE in 87% of
patients, although the majority of these resolved within six
energy of ablation (88) and therefore presumably not necessarily months (98). These patients usually had a degree of residual
related to the transseptal puncture itself. The occurrence of mitral stenosis; thus, a 10-mm Hg or more driving pressure
tamponade underestimates the frequency of pericardial effusion across an iatrogenic 5- to 10-mm defect in the septum would
secondary to small iatrogenic effusions: in one study of result in a significant sized permanent ASD. In addition,
1150 patients with left atrial access preparatory to planned atrial balloon dilatation of the septum can cause ripping rather
fibrillation ablation, effusion occurred in 2.7% while tamponade than stretching of the septal opening. At present, dilators are
requiring pericardiocentesis occurred in approximately one-third typically used to create a sufficient size fenestration, even for
of these patients (1%) (89). An uncommon but important cause of devices larger than the typical 14 Fr used for mitral dilatation.
tamponade is stitch perforation (90). In this scenario, the needle Persistent ASD after electrophysiology procedures, particu-
exits the RA and enters the LA while traversing tissue folds or larly when two catheters are placed across the septum, is also
periadventitial fat (Fig. 23.18). The separation between the cham- seen in approximately 87% of patients, with resolution in all
ber walls is limited, and the operator is typically unaware of except 4% by one year (99). Using two transseptal punctures
extracardiac passage of the needle. During most of the subse- rather than placing two catheters across a single fenestration
quent procedure, tamponade may not occur because the fenes- appears to decrease the risk of a persistent shunt (100).
tration is sealed by catheters or sheaths; when these are removed Because the left-to-right gradient is smaller in electrophysiol-
at the end of the case, effusion and tamponade may occur. The ogy procedures, the presence of a residual shunt remains more
phenomenon is most likely to occur with a low stick. Some common with mitral valvuloplasty (85).
operators address this possibility by leaving a guidewire in Transseptal puncture in patients with high right-sided
place prior to withdrawal of all hardware from the septum and pressures increases the risk of right to left shunting and systemic
observing for a few minutes to detect any new pericardial effu- arterial desaturation. One scenario is the use of a transseptally
sion. If in fact this occurs, the guidewire can provide access for placed percutaneous ventricular assist device (TandemHeart) in
placing a sheath to tamponade the hole temporarily while anti- the setting of right heart failure or right ventricular infarction
coagulation is reversed (86). when a PFO is also present. The simultaneous decompression of
the LA has been demonstrated to result in substantial cyanosis
because the iatrogenic pressure gradient drives large amount of
Embolization blood across the PFO from right to left (101). For similar reasons,
Once the Mullins sheath has been placed, maintenance of in the rare case of percutaneous dilatation of both tricuspid and
adequate anticoagulation, proper flushing and avoidance of mitral stenosis, the tricuspid valve should be dilated first, to
introduction of air are integral to the safety of the procedure. prevent a setting in which left atrial pressure is substantially
[14/7/010/20:46:50] [275–296]
lower than right atrial pressure after mitral valve dilatation be fatal, one other alternative has been described: access to the
results in right to left shunting (102). pericardium using an intracardiac approach with the transseptal
apparatus purposely utilized to perforate the heart and enter the
pericardial space (105).
Bezold-Jarisch Reflex
ST segment elevation, accompanied by chest pain, hypotension,
bradycardia and diaphoresis has been reported by a number of Air or Clot Embolism
operators to occur during or immediately after transseptal Treatment of air embolization is variably successful. Aggressive
access. While the phenomenon resembles a Bezold-Jarisch reflex, oxygenation, particularly in a hyperbaric chamber, and espe-
the mechanism is unclear. It occurs in somewhat less than 1% of cially if performed promptly, is the procedure of choice (91).
cases (89). Although introduction of air with consequent right Infusion of volume to maintain cerebral perfusion and admin-
coronary embolization has been postulated (103), coronary istration of lidocaine to protect cerebral tissue may also be
angiography during such episodes has failed to confirm beneficial. Aggressive transcatheter suctioning of air if trapped
this (104), and a neurally mediated mechanism remains the likely in a cardiac chamber has occasionally prevented clinically
etiology. The phenomenon is typically transient, and may significant embolism. Treatment of clot embolization is more
respond to atropine. complex. Iatrogenic embolization of thrombus has been treated
by intravenous (106) and intraarterial (107) thrombolysis, as
well as a variety of clot disruption (108) and extraction techni-
TREATMENT OF COMPLICATIONS ques (107), although the evidence base remains minimal (109).
Tamponade We have successful experience with thrombus embolization
Prompt recognition of tamponade is essential to successful from a transseptal sheath to the left main coronary artery
outcome. If ICE or TEE is performed during the procedure, it treated with aspiration using techniques primarily employed
can provide early recognition of pericardial effusion in over for primary coronary intervention in acute myocardial infrac-
80% of cases prior to the onset of hemodynamic compromise tion (110).
(65). Hypotension is commonly the first sign, although early in
tamponade hypertension and tachycardia may occur, likely
secondary to catecholamine stimulation (86). In some patients NEW TECHNOLOGIES AND
there is an abrupt slowing of heart rate, likely a vasovagal FUTURE CONSIDERATIONS
response to sudden pericardial stretch. Knowledge of the To facilitate safe transseptal puncture and modernize the
patient’s baseline hemodynamics is helpful; incipient or actual Mullins apparatus, a transseptal access device, ACross
tamponade should be accompanied by familiar hemodynamic (St. Jude Medical), has been developed. The device prevents
findings, including a narrow pulse pressure with pulsus para- inadvertent advancement of the needle while still inside the
doxus, near-obliteration during inspiration, and elevation with dilator, a concept that had been described previously using a
equalization (in most cases) of right and left heart filling pres- safety stop (111). Several new technologies provide alterna-
sures. Early diagnosis of tamponade is greatly facilitated by tives to the Brockenbrough needle and are designed to require
continuous arterial pressure display during the procedure. One less force to cross the septum, avoiding the typical high
of the earliest and most specific findings in the catheterization velocity movement with a rigid needle and attendant risks
laboratory is straightening and immobility of the left heart border associated with uncontrolled advance of both needle and
that is readily seen on fluoroscopy in the anterior-posterior view. Mullins assembly. The techniques are particularly appealing
This reflects the profile of the tense pericardium along with the when there is a thick or fibrosed septum that may be difficult
markedly diminished stroke volume that accompanies tampo- or impossible to penetrate. One of the technologies uses radio-
nade. Although ICE or TEE can provide immediate confirmation frequency energy to puncture the septum and includes a
of the diagnosis, if the procedure is done without continuous catheter system (112) that is approved by the U.S. Food and
echo guidance, a transthoracic echo can be obtained. However, if Drug Administration (Baylis Medical, Montreal, Quebec,
the echocardiographic equipment is not already in the cardiac Canada). The system has a curved and flexible tip designed
catheterization laboratory, waiting for its arrival prior to perfor- to decrease the risk of trauma to the LA, and sideholes to
mance of pericardiocentesis may be fatal and the operator needs facilitate dye injection. Using a similar approach, radiofre-
to proceed promptly in the setting of critical hemodynamic quency has been applied directly through a transseptal needle
compromise. Once the pericardium has been tapped, a catheter to facilitate transseptal puncture (113).
should be left in place for drainage; the average withdrawn after Another technique to minimize the force required to
tamponade in anticoagulated patients was greater than 800 mL in cross the septum and decrease perforation risk is a 0.014-in
one study of 15 periablation tamponades (65). Reversal of anti- nitinol wire with a needle like tip that assumes a J shape once
coagulation is helpful in this setting, and is one reason that we free in the LA and functions as a guidewire to facilitate dilator
routinely use heparin rather than bivalirudin in patients under- access (114) (SafeSept, Pressure Products, San Pedro, California,
going transseptal puncture. Surgical evacuation of the pericar- U.S.). Two other technologies should be mentioned. One exper-
dium is usually not necessary, frequently results in the surgeon imental device under development is the LACross (St. Jude
finding no obvious locus of perforation, and may expose the Medical). It features a sheath placed in the RA either from the
patient to substantial unnecessary morbidity and some mortality. internal jugular or femoral vein, with a sidehole that allows a
Nevertheless, in certain scenarios, such as inadvertent laceration catheter with a hollow screw to drill through the septum. Once
of a coronary artery, there may be no alternative to surgery (65). If in the LA, an Inoue wire can be advanced allowing atraumatic
the patient is in extremis, transthoracic pericardiocentesis is introduction of additional catheters. Finally, laser catheters
unsuccessful, and the delay to surgical evacuation is likely to have been used for controlled access to the LA, both in an
[14/7/010/20:46:50] [275–296]
animal model using magnetic resonance guidance (57) and in a diocentesis have been proposed (117) as minimal requirements.
small series of patients (115) (Fig. 23.19). Transseptal puncture meets the criteria for procedures where
Training for competence to perform transseptal punctu- simulator training is appealing: high-risk, relative low volumes
res has been challenging. There is no consensus on the number of procedures to which most trainees are exposed, and poten-
of procedures required to achieve or maintain competence tial for additive risk to the patient when the procedure is done
despite the significant learning curve and periprocedural under supervision rather than entirely by an experienced pri-
risks associated with the procedure. Training with a minimum mary operator. Accordingly, several simulation systems have
of 20 transseptals (116) and proficiency at emergency pericar- been developed.
Figure 23.19 A gallery of novel technologies for transseptal puncture: (A) The A Cross device is a Mullins apparatus with needle and safety
lock (St. Jude Medical, St. Paul, Minnesota, U.S.); (B) the Toronto catheter is designed to allow controlled radiofrequency entry, contrast
injection and sheath introduction (Baylis Medical, Montreal, Quebec, Canada); (C) radiofrequency applied through a transseptal needle;
(D) magnetic resonance imaging used for laser transseptal access in an animal model; (E) a nitinol needle system (SafeSept, Pressure
Products, San Pedro, California, U.S.); (F) the LACross (St. Jude Medical) is designed to allow transseptal puncture with a catheter deployed
through a side fenestration of a sheath placed in the RA. Technologies in (C–D) and (F) are experimental or offlabel. Abbreviations: LA, left
atrium; RA, right atrium; LAA, left atrial appendage. Source: Courtesy of the manufacturers (A–B and E–F) and the publishers (C–D) (see text
for manufacturer name and location), Refs. 57 and 111–113.
[14/7/010/20:46:50] [275–296]
Figure 23.19 Continued
CONCLUSIONS 8. O’Keefe JH Jr. Vlietstra RE, Hanley PC, et al. Revival of the
Transseptal puncture, the gateway to the LA for a growing transseptal approach for catheterization of the left atrium and
variety of structural heart disease and electophysiologic proce- ventricle. Mayo Clin Proc 1985; 60:790–795.
dures, is of increasing importance in invasive cardiology. The 9. Geske JB, Sorajja P, Ommen SR, et al. Left ventricular outflow
tract gradient variability in hypertrophic cardiomyopathy. Clin
significant benefits of proficiency in performing transseptals is
Cardiol 2009; 32:397–402.
somewhat offset by the substantial learning curve and compli- 10. De Ponti R, Cappato R, Curnis A, et al. Trans-septal catheterization
cation profile associated with the procedure. However, the in the electrophysiology laboratory: data from a multicenter survey
dramatic rise in number of transseptals performed has been spanning 12 years. J Am Coll Cardiol 2006; 47: 1037–1042.
accompanied by an expanding evidence base in the literature 11. Mullins CE. Transseptal left heart catheterization: experience
addressing issues such as optimal periprocedure management, with a new technique in 520 pediatric and adult patients. Pediatr
in particular anticoagulation. A number of technologies, includ- Cardiol 1983; 4:239–245.
ing adjunctive imaging and, more recently, equipment to facil- 12. Babaliaros VC, Green JT, Lerakis S, et al. Emerging applications
itate the transseptal puncture itself, may improve the overall for transseptal left heart catheterization old techniques for new
success rate and safety of this now half-century-old procedure. procedures. J Am Coll Cardiol 2008; 51:2116–2122.
13. Chaudhari RG, Shinde SV, Deshpande JR. Morphometric analy-
sis of fossa ovalis in rheumatic heart disease. Indian Heart J 2005;
57:662–665.
REFERENCES 14. Arya A, Hindricks G, Sommer P, et al. Long-term results and the
1. Cope C. Technique for transseptal catheterization of the left predictors of outcome of catheter ablation of atrial fibrillation
atrium; preliminary report. J Thorac Surg 1959; 37:482–486. using steerable sheath catheter navigation after single procedure
2. Bailey CP. The surgical treatment of mitral stenosis (mitral in 674 patients. Europace 2010; 12:173–180.
commissurotomy). Dis Chest 1949; 15:377–396. 15. Sorajja P, Cabalka AK, Hagler DJ, et al. Successful percutaneous
3. Ross J Jr., Braunwald E, Morrow AG. Transseptal left atrial repair of perivalvular prosthetic regurgitation. Catheter Cardio-
puncture; new technique for the measurement of left atrial pres- vasc Interv 2007; 70:815–823.
sure in man. Am J Cardiol 1959; 3:653–655. 16. Holmes DR Jr., Monahan KH, Packer D. Pulmonary vein stenosis
4. Russell RO Jr., Carroll JF, Hood WG Jr. Cardiac tamponade: a complicating ablation for atrial fibrillation: clinical spectrum and
complication of the transseptal technic of left heart catheter- interventional considerations. JACC Cardiovasc Interv 2009;
ization resulting in a fatality. Am J Cardiol 1964; 13:558–563. 2:267–276.
5. Dexter L, Haynes FW, Burwell CS, et al. Studies of congenital 17. Turi ZG. Percutaneous balloon valvuloplasty. In: Kern MJ, ed.
heart disease. II. The pressure and oxygen content of blood in the Hemodynamic Rounds: Interpretation of Cardiac Pathophysiol-
right auricle, right ventricle, and pulmonary artery in control ogy from Pressure Waveform Analysis. New York: Wiley-Liss,
patients, with observations on the oxygen saturation and source 2009:173–196.
of pulmonary “capillary” blood. J Clin Invest 1947; 26:554–560. 18. Schoenfeld MH, Palacios IF, Hutter AM Jr., et al. Underestima-
6. Swan HJ, Ganz W, Forrester J, et al. Catheterization of the heart tion of prosthetic mitral valve areas: role of transseptal catheter-
in man with use of a flow-directed balloon-tipped catheter. ization in avoiding unnecessary repeat mitral valve surgery.
N Engl J Med 1970; 283:447–451. J Am Coll Cardiol 1985; 5:1387–1392.
7. Schoonmaker FW, Vijay NK, Jantz RD. Left atrial and ventricular 19. Kober G, Hilgermann R. Catheter entrapment in a Bjork-Shiley
transseptal catheterization review: losing skills. Cathet Cardio- prosthesis in aortic position. Cathet Cardiovasc Diagn 1987; 13:
vasc Diagn 1987; 13:233–238. 262–265.
[14/7/010/20:46:50] [275–296]
20. Carabello BA, Barry WH, Grossman W. Changes in arterial 39. Eisenhauer AC, Hadjipetrou P, Piemonte TC. Balloon aortic valvu-
pressure during left heart pullback in patients with aortic loplasty revisited: the role of the inoue balloon and transseptal
stenosis: a sign of severe aortic stenosis. Am J Cardiol 1979; antegrade approach. Catheter Cardiovasc Interv 2000; 50:484–491.
44:424–427. 40. Sakata Y, Syed Z, Salinger MH, et al. Percutaneous balloon aortic
21. Gordon JB, Folland ED. Analysis of aortic valve gradients by valvuloplasty: antegrade transseptal vs. conventional retrograde
transseptal technique: implications for noninvasive evaluation. transarterial approach. Catheter Cardiovasc Interv 2005; 64:314–321.
Cathet Cardiovasc Diagn 1989; 17:144–151. 41. Cribier A, Eltchaninoff H, Tron C, et al. Early experience with
22. Carabello BA. Advances in the hemodynamic assessment of percutaneous transcatheter implantation of heart valve prosthesis
stenotic cardiac valves. J Am Coll Cardiol 1987; 10:912–919. for the treatment of end-stage inoperable patients with calcific
23. Omran H, Schmidt H, Hackenbroch M, et al. Silent and apparent aortic stenosis. J Am Coll Cardiol 2004; 43:698–703.
cerebral embolism after retrograde catheterisation of the aortic 42. Hanzel GS, Harrity PJ, Schreiber TL, et al. Retrograde percuta-
valve in valvular stenosis: a prospective, randomised study. neous aortic valve implantation for critical aortic stenosis. Cath-
Lancet 2003; 361:1241–1246. eter Cardiovasc Interv 2005; 64:322–326.
24. Hamon M, Gomes S, Oppenheim C, et al. Cerebral microembo- 43. Aragon J, Lee MS, Kar S, et al. Percutaneous left ventricular assist
lism during cardiac catheterization and risk of acute brain injury: device: “TandemHeart” for high-risk coronary intervention.
a prospective diffusion-weighted magnetic resonance imaging Catheter Cardiovasc Interv 2005; 65:346–352.
study. Stroke 2006; 37:2035–2038. 44. Hussein AA, Martin DO, Saliba W, et al. Radiofrequency ablation of
25. Jayne JE, Catherwood E, Niles NW, et al. Double-lumen catheter atrial fibrillation under therapeutic international normalized ratio:
assessment of aortic stenosis: comparison with separate catheter a safe and efficacious periprocedural anticoagulation strategy.
technique. Cathet Cardiovasc Diagn 1993; 29:157–160. Heart Rhythm 2009; 6:1425–1429.
26. Bae JH, Lerman A, Yang E, et al. Feasibility of a pressure wire 45. Marcus GM, Ren X, Tseng ZH, et al. Repeat transseptal catheter-
and single arterial puncture for assessing aortic valve area in ization after ablation for atrial fibrillation. J Cardiovasc Electro-
patients with aortic stenosis. J Invasive Cardiol 2006; 18:359–362. physiol 2007; 18:55–59.
27. Ross J Jr. Transseptal left heart catheterization a 50-year odyssey. 46. Lakkireddy D, Rangisetty U, Prasad S, et al. Intracardiac echo-
J Am Coll Cardiol 2008; 51:2107–2115. guided radiofrequency catheter ablation of atrial fibrillation in
28. Pearce AC, Schwengel RH, Simione LM, et al. Antegrade selec- patients with atrial septal defect or patent foramen ovale repair: a
tive coronary angiography via the transseptal approach in a feasibility, safety, and efficacy study. J Cardiovasc Electrophysiol
patient with severe vascular disease. Cathet Cardiovasc Diagn 2008; 19:1137–1142.
1992; 26:300–303. 47. Blackshear JL, Odell JA. Appendage obliteration to reduce stroke
29. Joseph G, Krishnaswami S, Baruah DK, et al. Transseptal in cardiac surgical patients with atrial fibrillation. Ann Thorac
approach to aortography and carotid artery stenting in pulseless Surg 1996; 61:755–759.
disease. Cathet Cardiovasc Diagn 1997; 40:416–420. 48. Chen WJ, Chen MF, Liau CS, et al. Safety of percutaneous
30. Yoon YS, Shim WH. Transseptal approach for stent implantation transvenous balloon mitral commissurotomy in patients with
in right internal carotid artery stenosis. J Invasive Cardiol 2000; mitral stenosis and thrombus in the left atrial appendage. Am J
12:70–74. Cardiol 1992; 70:117–119.
31. Montenero AS, Crea F, Bendini MG, et al. Catheter ablation of left 49. Cafri C, de la Guardia B, Barasch E, et al. Transseptal puncture
accessory atrioventricular connections: the transseptal approach. guided by intracardiac echocardiography during percutaneous
J Interv Cardiol 1995; 8:806–812. transvenous mitral commissurotomy in patients with distorted
32. Bonanno C, Paccanaro M, La Vecchia L, et al. Efficacy and safety anatomy of the fossa ovalis. Catheter Cardiovasc Interv 2000;
of catheter ablation versus antiarrhythmic drugs for atrial fibril- 50:463–467.
lation: a meta-analysis of randomized trials. J Cardiovasc Med 50. Hung JS, Fu M, Yeh KH, et al. Usefulness of intracardiac
(Hagerstown), 2009:408–818. echocardiography in complex transseptal catheterization during
33. Inoue K, Kitamura F, Chikusa H, et al. Atrial septostomy by a percutaneous transvenous mitral commissurotomy. Mayo Clin
new balloon catheter. Jpn Circ J 1981; 45:730–738. Proc 1996; 71:134–140.
34. Bonow RO, Carabello BA, Chatterjee K, et al. ACC/AHA 2006 51. Hurrell DG, Nishimura RA, Symanski JD, et al. Echocardiogra-
guidelines for the management of patients with valvular heart phy in the invasive laboratory: utility of two-dimensional echo-
disease: a report of the American College of Cardiology/Amer- cardiography in performing transseptal catheterization. Mayo
ican Heart Association Task Force on practice guidelines (writing Clin Proc 1998; 73:126–131.
committee to revise the 1998 guidelines for the management of 52. Patel A, Au E, Donegan K, et al. Multidetector row computed
patients with valvular heart disease) developed in collaboration tomography for identification of left atrial appendage filling
with the Society of Cardiovascular Anesthesiologists endorsed defects in patients undergoing pulmonary vein isolation for
by the Society for Cardiovascular Angiography and Interven- treatment of atrial fibrillation: comparison with transesophageal
tions and the Society of Thoracic Surgeons. J Am Coll Cardiol echocardiography. Heart Rhythm 2008; 5:253–260.
2006; 48:e1–e148. 53. Knight BP. Transesophageal echocardiography before atrial
35. Feldman T, Kar S, Rinaldi M, et al. Percutaneous mitral repair fibrillation ablation: looking before cooking. J Am Coll Cardiol
with the MitraClip system: safety and midterm durability in the 2009; 54:2040–2042.
initial EVEREST (Endovascular Valve Edge-to-Edge REpair 54. Calkins H, Brugada J, Packer DL, et al. HRS/EHRA/ECAS
Study) cohort. J Am Coll Cardiol 2009; 54:686–694. expert Consensus Statement on catheter and surgical ablation
36. Holmes DR, Reddy VY, Turi ZG, et al. Percutaneous closure of of atrial fibrillation: recommendations for personnel, policy,
the left atrial appendage versus warfarin therapy for prevention procedures and follow-up. A report of the Heart Rhythm Society
of stroke in patients with atrial fibrillation: a randomised non- (HRS) Task Force on catheter and surgical ablation of atrial
inferiority trial. Lancet 2009; 374:534–542. fibrillation. Heart Rhythm 2007; 4:816–861.
37. Ritzema J, Melton IC, Richards AM, et al. Direct left atrial 55. Chierchia GB, Capulzini L, de Asmundis C, et al. First experience
pressure monitoring in ambulatory heart failure patients: initial with real-time three-dimensional transoesophageal echocardiog-
experience with a new permanent implantable device. Circula- raphy-guided transseptal in patients undergoing atrial fibrilla-
tion 2007; 116:2952–2959. tion ablation. Europace 2008; 10:1325–1328.
38. McMahon CJ, El Said HG, Mullins CE. Use of the transseptal 56. Graham LN, Melton IC, Macdonald S, et al. Value of CT local-
puncture in transcatheter closure of long tunnel-type patent ization of the fossa ovalis prior to transseptal left heart catheter-
foramen ovale. Heart 2002; 88:E3. ization for left atrial ablation. Europace 2007; 9:417–423.
[14/7/010/20:46:50] [275–296]
57. Elagha AA, Kocaturk O, Guttman MA, et al. Real-time MR 78. Okumura Y. Which is the optimal therapy to prevent throm-
imaging-guided laser atrial septal puncture in swine. J Vasc boembolism after atrial fibrillation ablation procedures in low
Interv Radiol 2008; 19:1347–1353. stroke risk patients, anticoagulation or antiplatelet therapy?
58. Kultursay H, Turkoglu C, Akin M, et al. Mitral balloon valvulo- J Cardiovasc Electrophysiol 2009; 20:994–996.
plasty with transesophageal echocardiography without using 79. Cha YM, Powell BD, Hammill SC. Prevention of thromboembolic
fluoroscopy. Cathet Cardiovasc Diagn 1992; 27:317–321. stroke in patients undergoing catheter-based ablation for atrial
59. Poirier P, Champagne J, Alain P, et al. Mitral balloon valvuloplasty fibrillation: has it been optimized? J Cardiovasc Electrophysiol
in pregnancy: limiting radiation and procedure time by using 2009; 20:1364–1365.
transesophageal echocardiography. Can J Cardiol 1997; 13:843–845. 80. Ren JF, Marchlinski FE, Callans DJ, et al. Increased intensity of
60. Mishra S, Narang R, Sharma M, et al. Percutaneous transseptal anticoagulation may reduce risk of thrombus during atrial fibril-
mitral commissurotomy in pregnant women with critical mitral lation ablation procedures in patients with spontaneous echo
stenosis. Indian Heart J 2001; 53:192–196. contrast. J Cardiovasc Electrophysiol 2005; 16:474–477.
61. Doorey AJ, Goldenberg EM. Transseptal catheterization in adults: 81. Kilicaslan F, Verma A, Saad E, et al. Transcranial Doppler
enhanced efficacy and safety by low-volume operators using a “non- detection of microembolic signals during pulmonary vein
standard” technique. Cathet Cardiovasc Diagn 1991; 22:239–243. antrum isolation: implications for titration of radiofrequency
62. Tzeis S, Andrikopoulos G, Deisenhofer I, et al. Transseptal energy. J Cardiovasc Electrophysiol 2006; 17:495–501.
catheterization: considerations and caveats. Pacing Clin Electro- 82. Franzen OW, Klemm H, Hamann F, et al. Mechanisms underly-
physiol 2009:231–242. ing air aspiration in patients undergoing left atrial catheteriza-
63. Kong XQ, Wang L, Xue YZ, et al. A new approach for transseptal tion. Catheter Cardiovasc Interv 2008; 71:553–558.
catheterization in patients undergoing percutaneous balloon 83. Rihal CS, Nishimura RA, Holmes DR Jr. Percutaneous balloon
mitral valvuloplasty. Cardiology 2002; 98:46–49. mitral valvuloplasty: the learning curve. Am Heart J 1991;
64. Park SH, Kim MA, Hyon MS. The advantages of On-line trans- 122:1750–1756.
esophageal echocardiography guide during percutaneous bal- 84. Roelke M, Smith AJ, Palacios IF. The technique and safety of
loon mitral valvuloplasty. J Am Soc Echocardiogr 2000; 13:26–34. transseptal left heart catheterization: the Massachusetts General
65. Bunch TJ, Asirvatham SJ, Friedman PA, et al. Outcomes after Hospital experience with 1,279 procedures. Cathet Cardiovasc
cardiac perforation during radiofrequency ablation of the atrium. Diagn 1994; 32:332–339.
J Cardiovasc Electrophysiol 2005; 16:1172–1179. 85. Liu TJ, Lai HC, Lee WL, et al. Immediate and late outcomes of
66. Bidoggia H, Maciel JP, Alvarez JA. Transseptal left heart cathe- patients undergoing transseptal left-sided heart catheterization
terization: usefulness of the intracavitary electrocardiogram in for symptomatic valvular and arrhythmic diseases. Am Heart J
the localization of the fossa ovalis. Cathet Cardiovasc Diagn 1991; 2006; 151:235–241.
24:221–225. 86. Holmes DR Jr., Nishimura R, Fountain R, et al. Iatrogenic
67. Hung JS. Atrial septal puncture technique in percutaneous transpericardial effusion and tamponade in the percutaneous intra-
venous mitral commissurotomy: mitral valvuloplasty using the cardiac intervention era. JACC Cardiovasc Interv 2009; 2:705–717.
Inoue balloon catheter technique. Cathet Cardiovasc Diagn 1992; 87. Glazier JJ, Turi ZG. Percutaneous balloon mitral valvuloplasty.
26:275–284. Prog Cardiovasc Dis 1997; 40:5–26.
68. Krishnamoorthy KM, Dash PK. Transseptal catheterization with- 88. Hsu LF, Jais P, Hocini M, et al. Incidence and prevention of
out needle puncture. Scand Cardiovasc J 2001; 35:199–200. cardiac tamponade complicating ablation for atrial fibrillation.
69. Cribier A, Eltchaninoff H, Koning R, et al. Percutaneous mechan- Pacing Clin Electrophysiol 2005; 28(suppl 1):S106–S109.
ical mitral commissurotomy with a newly designed metallic 89. Fagundes RL, Mantica M, de Luca L, et al. Safety of single
valvulotome: immediate results of the initial experience in 153 transseptal puncture for ablation of atrial fibrillation: retrospec-
patients. Circulation 1999; 99:793–799. tive study from a large cohort of patients. J Cardiovasc Electro-
70. Cheng A, Calkins H. A conservative approach to performing physiol 2007; 18:1277–1281.
transseptal punctures without the use of intracardiac echocar- 90. Deshpande J, Vaideeswar P, Sivaraman A, et al. Balloon mitral
diography: stepwise approach with real-time video clips. valvotomy: an autopsy study. Int J Cardiol 1995; 52:67–76.
J Cardiovasc Electrophysiol 2007; 18:686–689. 91. Mofrad P, Choucair W, Hulme P, et al. Case report: cerebral air
71. Haruta S, Kouno H, Akanuma H, et al. The guidewire technique embolization in the electrophysiology laboratory during trans-
for transseptal puncture. J Invasive Cardiol 2005; 17:68–70. septal catheterization: curative treatment of acute left hemipare-
72. Hildick-Smith D, McCready J, de Giovanni J. Transseptal punc- sis with prompt hyperbaric oxygen therapy. J Interv Card
ture: use of an angioplasty guidewire for enhanced safety. Cath- Electrophysiol 2006; 16:105–109.
eter Cardiovasc Interv 2007; 69:519–521. 92. Cauchemez B, Extramiana F, Cauchemez S, et al. High-flow
73. Bruce CJ, Friedman PA, Narayan O, et al. Early heparinization perfusion of sheaths for prevention of thromboembolic compli-
decreases the incidence of left atrial thrombi detected by intra- cations during complex catheter ablation in the left atrium.
cardiac echocardiography during radiofrequency ablation for J Cardiovasc Electrophysiol 2004; 15:276–283.
atrial fibrillation. J Interv Card Electrophysiol 2008; 22:211–219. 93. Maleki K, Mohammadi R, Hart D, et al. Intracardiac ultrasound
74. Wazni OM, Beheiry S, Fahmy T, et al. Atrial fibrillation ablation detection of thrombus on transseptal sheath: incidence, treatment,
in patients with therapeutic international normalized ratio: com- and prevention. J Cardiovasc Electrophysiol 2005; 16:561–565.
parison of strategies of anticoagulation management in the 94. Kok LC, Mangrum JM, Haines DE, et al. Cerebrovascular com-
periprocedural period. Circulation 2007; 116:2531–2534. plication associated with pulmonary vein ablation. J Cardiovasc
75. Cortes M, Garcia E, Garcia-Fernandez MA, et al. Usefulness of Electrophysiol 2002; 13:764–767.
transesophageal echocardiography in percutaneous transcatheter 95. Chilukuri K, Sinha S, Berger R, et al. Association of transseptal
repairs of paravalvular mitral regurgitation. Am J Cardiol 2008; punctures with isolated migraine aura in patients undergoing
101:382–386. catheter ablation of cardiac arrhythmias. J Cardiovasc Electro-
76. Cappato R, Calkins H, Chen SA, et al. Prevalence and causes of physiol, 2009:1227–1230 .
fatal outcome in catheter ablation of atrial fibrillation. J Am Coll 96. Kessler DJ, Pirwitz MJ, Horton RP, et al. Intracardiac shunts
Cardiol 2009; 53:1798–1803. resulting from transseptal catheterization for ablation of acces-
77. Dagres N, Hindricks G, Kottkamp H, et al. Complications of sory pathways in otherwise normal hearts. Am J Cardiol 1998;
atrial fibrillation ablation in a high-volume center in 1,000 82:391–392.
procedures: still cause for concern? J Cardiovasc Electrophysiol 97. Thomas MR, Monaghan MJ, Metcalfe JM, et al. Residual atrial
2009; 20:1014–1019. septal defects following balloon mitral valvuloplasty using
[14/7/010/20:46:50] [275–296]
different techniques. A transthoracic and transoesophageal echo- 108. Chan AW, Henderson MA. Immediate catheter-directed reper-
cardiography study demonstrating an advantage of the Inoue fusion for acute stroke occurring during diagnostic cardiac
balloon. Eur Heart J 1992; 13:496–502. catheterization. Catheter Cardiovasc Interv 2006; 67:314–318.
98. Yoshida K, Yoshikawa J, Akasaka T, et al. Assessment of left-to- 109. Sankaranarayanan R, Msairi A, Davis GK. Stroke complicating
right atrial shunting after percutaneous mitral valvuloplasty by cardiac catheterization–a preventable and treatable complication.
transesophageal color Doppler flow-mapping. Circulation 1989; J Invasive Cardiol 2007; 19:40–45.
80:1521–1526. 110. Svilaas T, Vlaar PJ, van der Horst IC, et al. Thrombus aspiration
99. Rillig A, Meyerfeldt U, Birkemeyer R, et al. Persistent iatrogenic during primary percutaneous coronary intervention. N Engl J
atrial septal defect after pulmonary vein isolation: incidence and Med 2008; 358:557–567.
clinical implications. J Interv Card Electrophysiol 2008; 22:177–181. 111. Sakata Y, Feldman T. Transcatheter creation of atrial septal
100. Hammerstingl C, Lickfett L, Jeong KM, et al. Persistence of perforation using a radiofrequency transseptal system: novel
iatrogenic atrial septal defect after pulmonary vein isolation–an approach as an alternative to transseptal needle puncture. Cath-
underestimated risk? Am Heart J 2006; 152:362–365. eter Cardiovasc Interv 2005; 64:327–332.
101. Loyalka P, Idelchik GM, Kar B. Percutaneous left ventricular 112. Bidart C, Vaseghi M, Cesario DA, et al. Radiofrequency current
assist device complicated by a patent foramen ovale: importance delivery via transseptal needle to facilitate septal puncture. Heart
of identification and management. Catheter Cardiovasc Interv Rhythm 2007; 4:1573–1576.
2007; 70:383–386. 113. de Asmundis C, Chierchia GB, Sarkozy A, et al. Novel trans-
102. Sharma S, Loya YS, Desai DM, et al. Percutaneous double-valve septal approach using a Safe Sept J-shaped guidewire in difficult
balloon valvotomy for multivalve stenosis: immediate results left atrial access during atrial fibrillation ablation. Europace 2009;
and intermediate-term follow-up. Am Heart J 1997; 133:64–70. 11:657–659.
103. Turi ZG. Puncturing the septum: resurgent technique with inher- 114. Bloomfield DA. Transseptal catheterization technique: a simple
ent risk. J Invasive Cardiol 2004; 16:3–4. safety procedure. Cathet Cardiovasc Diagn 1991; 22:153–155.
104. Hildick-Smith DJ, Ludman PF, Shapiro LM. Inferior ST-segment 115. Weber HP. Transseptal Laser Puncture. Available at: http://www.
elevation following transseptal puncture for balloon mitral ccep. de/ablation/documents/en/Transseptal pdf. Accessed 2009.
valvuloplasty is atropine-responsive. J Invasive Cardiol 2004; 116. Linker NJ, Fitzpatrick AP. The transseptal approach for ablation
16:1–2. of cardiac arrhythmias: experience of 104 procedures. Heart 1998;
105. Hsu LF, Scavee C, Jais P, et al. Transcardiac pericardiocentesis: 79:379–382.
an emergency life-saving technique for cardiac tamponade. 117. Tracy CM, Akhtar M, DiMarco JP, et al. American College of
J Cardiovasc Electrophysiol 2003; 14:1001–1003. Cardiology/American Heart Association 2006 update of the
106. Serry R, Tsimikas S, Imbesi SG, et al. Treatment of ischemic clinical competence statement on invasive electrophysiology
stroke complicating cardiac catheterization with systemic throm- studies, catheter ablation, and cardioversion: a report of the
bolytic therapy. Catheter Cardiovasc Interv 2005; 66:364–368. American College of Cardiology/American Heart Association/
107. Al Mubarak N, Vitek JJ, Mousa I, et al. Immediate catheter-based American College of Physicians Task Force on clinical compe-
neurovascular rescue for acute stroke complicating coronary tence and training developed in collaboration with the Heart
procedures. Am J Cardiol 2002; 90:173–176. Rhythm Society. J Am Coll Cardiol 2006; 48:1503–1517.
[2/7/010/8:56:54] [297–306]
24
Mesenteric and renal angiography

Steven J. Filby, Mehdi H. Shishehbor, and Leslie Cho
INTRODUCTION disease. Renal arterial disease is prevalent in patients with

Catheter-based contrast angiography remains the “gold stan- other forms of atherosclerotic disease such as multivessel cor-
dard” in the diagnosis of mesenteric and renal artery disease onary artery disease and peripheral arterial disease. The clinical
and usually follows a high index of clinical suspicion and clues suggesting the diagnosis of RAS are detailed in Table 24.3.
corroborative noninvasive testing. Duplex ultrasonography, Of these, the strongest clinical indicators are early onset of
computed tomography angiography (CTA), and magnetic res- hypertension prior to age of 30 years; multidrug-resistant
onance angiography (MRA) have become extremely helpful in hypertension; development of renal failure or declining renal
the initial assessment of patients with suspected disease, but function after institution of angiotensin-converting enzyme
invasive angiography is often required for definitive diagnosis. inhibitor (ACE-I) or angiotensin receptor blocker (ARB); atro-
Catheter-based angiography has the added benefit of providing phic kidney or 1.5-cm size discrepancy between kidneys; flash
simultaneous interventional therapy if required. This chapter pulmonary edema; and unexplained renal failure. When renal
reviews the vascular anatomy and equipment essential to con- arterial disease is suspected, noninvasive imaging with duplex
ducting a proper mesenteric and renal angiographic examination ultrasound, MRA with gadolinium enhancement, or CTA is
and highlights the proper indications for invasive testing of these recommended to first establish the diagnosis of RAS.
vessels. Percutaneous mesenteric and renal interventions are
addressed in separate chapters.
CLINICAL ASPECTS
Acute Mesenteric Ischemia
INDICATIONS Although relatively infrequent, acute mesenteric ischemia is a
Mesenteric Angiography life-threatening condition caused by a sudden decrease in
The principal indication for mesenteric angiography is evalua- blood flow to the intestines. Prolonged intestinal hypoperfusion
tion of intestinal ischemia that may be acute or chronic (1). can culminate in bowel necrosis and death, and mesenteric
Classically, patients with acute intestinal ischemia present with ischemia is associated with mortality ranging from 60% to 100%
sudden or recent onset pain and little or no findings on (2,3). While the early use of angiography has resulted in a
abdominal physical exam. These patients often have a history decline in mortality rates over the past 30 years, patient out-
of coronary artery disease, peripheral arterial disease, or other come is highly dependent on prompt recognition and early
cardiovascular risk factors (Table 24.1). Acute intestinal ische- treatment of this disease process (4).
mia should also be suspected in patients with severe abdominal Acute mesenteric ischemia is characterized with dura-
pain following endovascular procedures that involve catheter tion of hours to days and may be caused by embolic, throm-
traversal in the abdominal aorta. The decision to perform botic, or nonocclusive etiologies (Table 24.1) (5,6). Classically,
angiography on patients with suspected acute intestinal ische- patients present with sudden severe abdominal pain out of
mia needs to be individualized. For instance, patients present- proportion to the findings on clinical exam and may have fever
ing with acute abdominal pain due to suspected arterial and bloody stool. With occlusive acute mesenteric ischemia,
occlusion with intestinal infarction should be referred for patients often have coexisting atherosclerotic disease or risk
immediate laparotomy instead of angiography. Alternatively, factors that predispose them to embolism or thrombus forma-
patients with nonocclusive disease may benefit from a strategy tion. Most often, occlusive acute mesenteric ischemia involves
that incorporates initial angiography. the proximal superior mesenteric artery (SMA) either by
As in those with acute ischemia, patients with chronic thrombosis of preexisting plaque or by embolism to this site
mesenteric ischemia often have coexisting atherosclerotic disease (7). In nonocclusive mesenteric ischemia, intestinal hypoper-
(Table 24.2). In cases of suspected chronic intestinal ischemia, fusion results from a low-flow state in the splanchnic circula-
there is often time for initial assessment with noninvasive testing tion caused by a sudden drop in systemic blood pressure,
including duplex ultrasound, CTA, or MRA. Angiographic eval- severe arterial vasospasm (e.g., from cocaine use or treatment
uation should be performed in patients with abnormal or with vasopressin or norepinephrine), or cardiogenic shock
indeterminate noninvasive testing. Alternatively, diagnostic mes- (2,6,8). The splenic flexure, a watershed area supplied by the
enteric angiography may be performed as an initial diagnostic SMA and inferior mesenteric artery (IMA), is particularly
tool if these noninvasive testing modalities are not available. susceptible to ischemia in low-flow states. Overall, the inci-
dence of acute mesenteric ischemia is rising (3). The reasons
Renal Angiography for this are unclear, but may in part be due to a growing
Renal angiography is indicated in patients in whom renal elderly population with atherosclerotic disease and increased
artery stenosis (RAS) is clinically suspected and who have proportion of patients with shock states requiring vasoactive
also corroborative noninvasive testing suggesting significant medications (3).
[2/7/010/8:56:54] [297–306]
Table 24.1 Risk Factors for the Development of Acute Mesenteric vessel is generally well tolerated. Patients who have had sur-
Ischemia gical interruption of these splanchnic collateral networks are
l
more likely to be symptomatic from single-vessel mesenteric
Advanced age
l
disease (10,11). Additionally, single-vessel disease involving
Low cardiac output
l Cardiac arrhythmia (e.g., atrial fibrillation) the SMA has been reported to cause chronic mesenteric ischemia.
l Valvular heart disease Chronic mesenteric ischemia can also result from mesenteric
l Recent myocardial infarction vein thrombosis, most often involving the superior mesenteric
l Malignancy vein. Primary venous thrombosis is caused by hereditary hyper-
l Hypercoagulable states coagulable states, while secondary venous thrombosis results
l Critical illness or prolonged intensive care unit stay from etiologies associated with malignancy or inflammation.
l Trauma While the mortality rate in chronic mesenteric ischemia is far
lower than that of acute mesenteric ischemia, the natural history
of chronic mesenteric ischemia is less well defined and some
Table 24.2 Risk Factors for the Development of Chronic Mesen- patients with chronic ischemia progress to acute mesenteric
teric Ischemia ischemia (12).
l Peripheral arterial disease
l Coronary artery disease Median Arcuate Ligament Syndrome
l Advanced age Median arcuate ligament syndrome (also known as celiac artery
l Diabetes mellitus compression syndrome, celiac axis syndrome, and Dunbar syn-
l Smoking drome) is a rare clinical entity thought to be caused by the
l Obesity compression of the celiac axis by anomalous fibrous diaphrag-
l Hyperlipidemia matic bands. It is characterized by postprandial abdominal
l Sedentary lifestyle discomfort and weight loss and is occasionally associated with
an abdominal bruit (13). While the pathophysiology of this
disorder is not clearly understood, some have suggested a con-
Table 24.3 Clinical Clues to the Diagnosis of Renal Artery Stenosis genital origin (14). Since the SMA and the IMA remain widely
patent in these cases, there should, in theory, still be ample blood
l Early onset of hypertension before the age of 30 yr supply to the bowel. As such, the diagnosis of median arcuate
l Onset of severe hypertension after the age of 55 yr
l
ligament syndrome remains controversial and is often one of
Multidrug-resistant hypertension
l
exclusion (15–17). Expiration may accentuate regional stricture in
Development of new renal failure or worsening renal function
following initiation of ACE (angiotensin converting enzyme) patients with median arcuate ligament syndrome, and catheter
inhibitor or ARB (angiotensin receptor blocker) therapy angiography during respiratory maneuvers is often helpful in
l Size difference between kidneys of more than 1.5 cm or evaluating these patients.
unexplained atrophic kidney
l
l
Systolic-diastolic epigastric bruit Visceral Artery Aneurysms
Unexplained renal failure
l
Aneurysms of the mesenteric vessels are uncommon. The
Multivessel coronary artery disease
l Unexplained or recurrent pulmonary edema
majority of patients with these visceral aneurysms are asymp-
l Angina refractory to medical management tomatic, and these aneurysms are found incidentally during
unrelated abdominal imaging (18,19). However, these aneur-
ysms do carry a risk of rupture and hemorrhage that can be
fatal. Splenic artery aneurysms are the most common occurring
Chronic Mesenteric Ischemia in 60% of cases. In one series, fewer than 20% of patients with
In comparison to acute mesenteric ischemia, chronic mesenteric splenic artery aneurysms presented with abdominal pain or
ischemia is more indolent in nature, with duration of onset over rupture (20). The mortality rate for nonpregnant patients with
weeks to months. Patients are most commonly female and splenic artery aneurysms ranges between 10% to 25% but may
present with abdominal pain, weight loss, and no alternative be as high as 70% for pregnant females (21). In general, splenic
explanation. The abdominal discomfort is variable in character aneurysms greater than 2.0 cm in diameter are thought to be of
and may have associated bloating; it typically occurs 30 minutes sufficient risk of rupture to warrant treatment. Aneurysms
to 3 hours after food ingestion. Recurrent intestinal angina involving the hepatic artery are increasingly common and
causes sitophobia ( fear of food). Patients suffering from chronic presently make up 20% of cases. This is probably due to an
intestinal ischemia often reduce their food intake and on an increase in percutaneous biliary procedures being performed
average lose 10 to 20 kg prior to diagnosis. today as well as increased recognition from incidental imaging
The patients are typically smokers and have additional (22). Aneurysms of the SMA are less common accounting for
cardiovascular risk factors or evidence of systemic atheroscle- only 6% of total cases. Aneurysms of the renal arteries are
rosis (Table 24.2). Indeed, progressive atherosclerotic disease most commonly associated with FMD, which is discussed in
causes the vast majority of chronic mesenteric ischemia (9). Less the following section. Vasculitis and trauma have also been
common etiologies include vasculitis, fibromuscular dysplasia associated with renal artery aneurysms (23).
(FMD), and aortic dissection. The classical teaching has been
that chronic mesenteric ischemia is nearly always caused by Renal Artery Stenosis
narrowing of two or more mesenteric vessels. The mesenteric Renovascular disease is well recognized as a potentially revers-
vessels—celiac trunk, SMA, and IMA—can develop collaterals ible cause of hypertension and renal failure. RAS has been
at multiple levels such that a high-grade stenosis of any single associated with coronary artery disease and congestive heart
[2/7/010/8:56:54] [297–306]
MESENTERIC AND RENAL ANGIOGRAPHY 299
failure, and patients with RAS have a markedly reduced sur- band-like stenoses often interposed with aneurysmal dilata-
vival rate. RAS by ultrasound is also prevalent in patients with tions. The angiographic appearance of FMD has thus been
other forms of atherosclerotic disease such as peripheral arterial characterized as “beads on a string” (Fig. 24.1). Unlike in
or coronary artery disease. Population-based studies have atherosclerotic RAS, distal or branch vessel involvement in
demonstrated that approximately 20% to 60% of patients with FMD may occur. Disease progression occurs in more than
peripheral arterial disease also have RAS (24). In a study of 30% of patients in FMD (30).
patients who underwent screening aortography at the time
of coronary angiography, 4.8% had significant renal stenoses
of more than 75% narrowing (25). ANATOMIC CONSIDERATIONS
Clinically, RAS may present as uncontrolled hyperten- Mesenteric Circulation
sion, flash pulmonary edema, intolerance to ACE-I or ARB The mesenteric arteries arise from the anterior aspect of the
treatment, progressive renal deterioration, or refractory angina. lower thoracic and abdominal aorta. These vessels—the celiac
The presence of renal arterial disease does not necessarily trunk, SMA, and IMA—are responsible for the blood supply to
indicate that the patient’s hypertension or renal failure is caused all organs located within the abdominal cavity. The celiac trunk
by RAS, and outcomes in renal revascularization studies is the first major branch of the abdominal aorta and is an
have been discordant (26–28). When hypertension is attributed essential source of blood supply to the liver, stomach, and
to RAS, the term “renovascular hypertension” is commonly parts of the esophagus, spleen, duodenum, and pancreas. Its
used. Flash pulmonary edema or renal failure following origin from the anterior aorta is typically midline at the level of
administration of ACE-I or ARB usually indicates significant the T12 vertebral body; and it courses inferiorly for 1 to 2 cm
bilateral renal arterial disease or significant disease in a soli- before branching into the left gastric, common hepatic, and
tary kidney. splenic arteries (Fig. 24.2). The common hepatic divides into
RAS is most commonly due to either atherosclerotic the proper hepatic artery and, typically, also the gastroduodenal
disease or FMD (29). Atherosclerosis is the main mechanism artery. The proper hepatic gives off the right gastric artery
of RAS in patients older than 55 years. Atherosclerotic renal before branching into the right and left hepatic arteries. The
arterial lesions are most often aorto-ostial in location, involving gastroduodenal artery then goes on to divide into the right
the ostial and proximal segment of the vessel; distal or branch gastroepiploic artery and the anterior and posterior superior
vessel involvement is uncommon (28). Progression in athero- pancreaticoduodenal arteries. The right gastroepiploic artery
sclerotic RAS occurs in more than 40% of patients (30). Fibro- and the left gastroepiploic artery (from the splenic artery) join
muscular dysplasia is seen predominately among young female together along the greater curvature of the stomach. The right
patients. The etiology of FMD is unknown, but a genetic pre- gastric artery and the left gastric artery join together to run along
disposition has been suggested (31). Degenerative changes lead the lesser curvature of the stomach. Because of the redundant
to fibroplasia and arterial wall weakening resulting in fibrous, blood supply to the stomach, gastric ischemia is uncommon.
Figure 24.1 Selective right renal angiogram demonstrating fibro- Figure 24.2 Normal anatomy of the celiac artery. (A) Common
muscular dysplasia (FMD). Degenerative changes lead to fibropla- hepatic artery; (B) left gastric artery; (C) esophageal branches;
sia and arterial wall weakening resulting in fibrous, band-like (D) splenic artery; (E) short gastric branches; (F) splenic branches;
stenoses often interposed with aneurysmal dilatations. The angio- (G) left gastroepiploic artery; (H) right gastric artery; (I) right
graphic appearance of FMD has thus been characterized as “beads gastroepiploic artery; (J) superior pancreaticoduodenal artery;
on a string” (arrow). (K) gastroduodenal artery; (L) hepatic artery. Source: From Ref. 32.
[2/7/010/8:56:54] [297–306]
Figure 24.4 Normal anatomy of the inferior mesenteric artery.

(A) left colic artery (superior and inferior branches); (B) lower left
colic artery; (C) sigmoid artery; (D) superior rectal artery. Source:
Figure 24.3 Normal anatomy of superior mesenteric artery. From Ref. 32.
(A) Inferior pancreaticoduodenal (anterior and posterior branches);
(B) middle colic artery; (C) jejunal artery; (D) ileal artery; (E) ileal
artery; (F) ileocolic artery; (G) right colic. Source: From Ref. 32.
effectively anastomosing the SMA and IMA circulations in
what is known as the arc of Riolan.
The SMA usually originates 1 cm lower than the celiac Renal Arteries
trunk and anterior to the L1 vertebral body. The SMA travels The renal arteries originate from the lateral abdominal aorta
inferiorly and slightly rightward to supply the duodenum and immediately below the SMA at the level of the lower border of
pancreas. In its course, the SMA passes beneath the pancreas the L1 vertebral body. There are slight anatomic differences
and divides into the inferior pancreaticoduodenal, middle colic, between the two renal arteries. In one study of 100 patients who
right colic, ileocolic, and intestinal branches (Fig. 24.3). In underwent spiral CTA, the right and left renal arteries origi-
general, the middle colic artery provides blood supply to nated at the same level in 50% of patients (33). In the other 50%
the proximal and mid-transverse colon. In some individuals, of cases, the right renal artery arose higher than the left. The
the middle colic may provide the main source of blood to the right renal artery also typically courses downward to the sup-
splenic flexure. The right colic artery provides the blood supply ply the more inferiorly located right kidney; the left renal artery
to the middle and distal ascending colon, while the ileocolic typically has a horizontal course. The right renal artery has an
artery supplies the distal ileum, cecum, and proximal ascend- anterolateral origin from the aorta while the left renal artery has
ing colon. The middle, right, and ileocecal branches join a posterolateral origin. Both renal arteries typically give off
together with the left colic artery (from the inferior mesenteric) small inferior suprarenal branches before dividing into seg-
forming the marginal artery or artery of Drummond that mental branches. These segmental branches subsequently
courses along the inside border of the colon. Multiple anatomic divide into arcuate and multiple interlobular branches termi-
variations of the colic arteries exist. nating within the renal cortex and medulla. Accessory renal
The IMA is the smallest of the mesenteric vessels. It arteries are the most common vascular variant (34). These
originates below the level of the renal arteries and approxi- accessory vessels may be of similar or smaller caliber and
mately 6 to 7 cm below the SMA. The IMA courses inferiorly typically originate lower than the main artery, supplying the
and leftward giving off the left colic artery and several sigmoid inferior pole of the kidney (Fig. 24.5). Another variant occurs
branches before terminating in the superior rectal artery when the main artery divides early in its course into segmental
(Fig. 24.4). The IMA is responsible for providing the blood branches. Finally, it should be noted that the renal arteries
supply to the distal transverse colon, descending colon, and the can be surgically bypassed or reimplanted in the pelvis
rectum. The left and middle colic branches may join together, (“autotransplant”).
[2/7/010/8:56:54] [297–306]
Figure 24.5 Accessory renal artery. Accessory renal arteries typically originate lower than the main artery and supply the inferior pole of the
kidney. In this angiogram, selective contrast injection of the main renal artery does not perfuse the inferior renal pole (A) as this segment is
instead perfused by an accessory artery (B). Fibromuscular dysplasia of the accessory artery is indicated by the arrow.
EQUIPMENT AND TECHNICAL CONSIDERATIONS The abdominal aortogram is first performed in a stan-
Abdominal Aortography dard posteroanterior (PA) projection using digital subtraction.
Prior to selective engagement, initial abdominal aortic angiog- Our practice is to use a 5-Fr system with a total volume of 10 to
raphy should be performed with a pigtail or Omniflush cath- 20 mL of contrast at a rate of 10 mL/sec. Patients should be
eter (Omni Flush, AngioDynamics Inc., Queensbury, New instructed not to breath or move prior to angiography. Lateral
York, U.S.) (Table 24.4). The Omniflush catheter is designed projection aortography may then be performed to visualize the
to minimize the amount of upward-refluxed contrast and thus mesenteric arteries since these vessels arise anteriorly
may result in more contrast concentrated at the level of the (Fig. 24.7). This should also be done with digital subtraction,
renal and mesenteric vessels. Usually, the aortogram is ade- and similar settings may be used. If the renals are the sole focus
quate to demonstrate patency but often is insufficient to ade- of the study, a 158 left anterior oblique (LAO) projection is
quately assess degree of stenosis of the mesenteric and renal recommended and can be performed using a smaller amount of
vessels. The aortogram does, however, provide important
adjunctive information regarding the presence of aortic calcifi-
cation, aneurysmal dilatation, and anatomical anomalies (e.g.,
number of accessory renal arteries) as well as the location of the
renal and mesenteric ostia (Fig. 24.6). The catheter should be
placed with its side holes placed at the T12-L1 intervertebral
space (i.e., above the origin of the renal arteries). For optimal
imaging, the field of view should be maximized, the table
elevated to its highest setting, and the table positioned such
that the catheter tip is at top of the screen. If a recent prior
abdominal aortogram is available for review, repeat study is
not necessary and the operator should proceed to selective
assessment of the arteries of interest.
Table 24.4 Equipment for Abdominal Aortic Angiography

l Sheath
l 4 Fr or 5 Fr
l Catheters
l Pigtail
l Straight Flush (AngioDynamics, Queensbury, New York, U.S.)
l Omniflush (AngioDynamics) Figure 24.6 Posteroanterior (PA) abdominal aortogram. (A) right
l Wires (0.035 in.) renal artery; (B) accessory right renal artery; (C) left renal artery;
l Standard J-wire (Cook Medical, Bloomington, Indiana, U.S.) (D) celiac trunk; (E) proper hepatic artery; (F) gastroduodenal artery.
l Glidewire (Terumo, Tokyo) The left renal artery has changes consistent with fibromuscular
l Wholey (Tyco Healthcare, Mallinckrodt, St. Louis, Missouri, U.S.) dysplasia (arrow).
l Rosen (Cook Medical)
[2/7/010/8:56:54] [297–306]
Figure 24.7 Lateral abdominal aortogram demonstrating the ante-

rior origins of the celiac artery (A) and superior mesenteric artery
(B). Nonobstructive narrowing of the proximal celiac artery (C) is
present. Figure 24.8 Selective angiography of the celiac trunk. As the first
major branch of the abdominal aorta, the celiac trunk (A) originates
from the anterior aorta at the level of the T12 vertebral body and
branches into the left gastric (B), common hepatic (C), and splenic
(D) arteries. The common hepatic artery further divides into the
contrast (10 mL/sec for 10 mL total). For these studies, the proper hepatic and gastroduodenal (E) arteries. This angiogram
contrast may be diluted as 70% dye and 30% heparinized demonstrates an anomalous left hepatic artery arising from the
saline. This dilution technique reduces the total contrast expo- gastroduodenal artery.
sure and still results in acceptable image quality.
Selective Mesenteric Angiography

A 5-Fr system is usually adequate for diagnostic mesenteric
angiography (Table 24.5). Femoral access is commonly used;
but in general, an arm approach is easier for the selective engage-
ment of the mesenteric vessels. From a femoral access, our pref-
erence is to use a reverse angulation catheter such as a Sos
(AngioDynamics) catheter that allows easy access to the inferiorly
directed ostia of the celiac trunk, SMA, and IMA (Figs. 24.8–24.11).
A JR 4 diagnostic, internal mammary artery or left coronary
bypass (LCB) catheter can also be used to engage these vessels.
Reverse angulation catheters should be positioned by moving
them caudally with a generous amount of wire extending from
the tip that allows for a relatively atraumatic bend of wire over
which the catheter can pass. Once the catheter is positioned
Table 24.5 Equipment for Selective Mesenteric Angiography

l Sheath
l 5 Fr or 6 Fr
l
Figure 24.9 Selective angiography of the superior mesenteric
Catheters
l JR 4
artery (SMA). The SMA (A) travels inferiorly and slightly rightward
l IMA
to supply the duodenum and pancreas. In its course, the SMA gives
l Sos
off the inferior pancreaticoduodenal arteries (B), the middle colic
l Simmons 1, 2, 3
artery (C), and several intestinal arteries (D).
l Cobra C2
l Hockey Stick
l Multipurpose (arm approach)
l Wires (0.035 in.) superiorly to the vessel ostia, its curve is formed by the slow
l Standard J-wire
withdrawal of the wire. If kept in an anterior orientation and
l Glidewire
walked down the aorta, such reverse angle catheters should
l Wholey
readily find the mesenteric ostia, but care should be taken as
l Rosen
these catheters also will readily catch onto sidewall atheroma.
[2/7/010/8:56:54] [297–306]
Figure 24.12 Selective angiography of the celiac trunk with respi-

ratory maneuver. In cases of suspected celiac artery compression,
comparative, selective views should be obtained at end-inspiration
and end-expiration. Expiration causes the aorta and its major
branches to move cephalad, thus exacerbating vessel compression
Figure 24.10 Selective angiography of the inferior mesenteric of the celiac artery by a median arcuate ligament. This angiogram
artery (IMA). The IMA (A) courses inferiorly and leftward giving off demonstrates moderate stenosis of the celiac trunk with expiration
the left colic artery (B), which in part supplies the marginal artery (arrow).
(C). In its terminal segment, the IMA supplies the sigmoid colon and
rectum (D). The arrow indicates an IMA stenosis prior to the origin of
the left colic branch.
Alternatively, many operators prefer an arm approach in

selective mesenteric studies. From a brachial or radial approach,
the mesenteric arteries can usually be easily engaged using a
multipurpose diagnostic catheter. In most cases catheters that
are 125 cm in length are necessary with the (right) radial
approach. Arm access is particularly helpful in patients with
severe aortoiliac tortuosity and highly angulated vessel origins.
Arm access should also be used in patients with known abdomi-
nal aortic aneurysms and patients with severe aortoiliac disease.
Selective angiography of the celiac trunk is performed
with a 158 to 308 LAO angulation to demonstrate the celiac axis
origin and trifurcation into the left gastric, common hepatic,
and splenic arteries. In cases of suspected celiac artery com-
pression, comparative, selective views should be obtained at
end-inspiration and end-expiration as expiration may exacer-
bate vessel compression (Fig 24.12). Given its course to the
rightward pelvis, angiography of the SMA should be per-
formed in a 158 to 308 LAO. In comparison, the IMA has a
leftward course; and thus, selective angiography of this vessel
should be performed in a 158 to 308 right anterior oblique
(RAO) projection. As with abdominal aortography, digital sub-
traction angiography is recommended. The field of view should
be adequate to visualize the mesenteric vessel of interest as well
as all potential collateral networks in cases of occlusion.
Figure 24.11 Selective angiography of the superior mesenteric

artery. The SMA and celiac vessels normally communicate via the Selective Renal Angiography
anterior and posterior pancreaticoduodenal arteries such that selec- When performing renal angiography, we predominately use
tive contrast injection of the SMA fills both vessels. (A) Common aortic femoral access with a 5-Fr system (Table 24.6). Arm access is
artery; (B) splenic artery; (C) gastroduodenal artery; (D) superior reserved for those patients with severe aortoiliac disease or
mesenteric artery; (E) posterior and anterior pancreaticoduodenal known abdominal aortic aneurysm. The JR 4 catheter is our
arteries; (F) intestinal arteries. workhorse catheter for selectively engaging the renal arteries.
Alternatively, an internal mammary or renal double curve
[2/7/010/8:56:54] [297–306]
Table 24.6 Equipment for Selective Renal Angiography

l Sheath
l 5 Fr or 6 Fr
l Catheters
l JR 4
l IMA
l Renal Double Curve
l Cobra C2
l Multipurpose (arm approach)
l Wires (0.035 in.)

l Standard J-wire
l Glidewire
l Wholey
l Rosen
Figure 24.14 Selective left renal angiogram (panels A and B).

Renal artery stenosis is an aorto-ostial disease process, and calci-
Figure 24.13 Selective angiography of the left renal artery using a fication can serve as an important landmark indicating the location
Cobra (C2) catheter. In cases of aortoiliac tortuosity, a C2 or of the renal ostia. The arrow in this angiogram indicates the pres-
Simmons catheter is sometimes necessary to reach the renal ence of circumferential calcium in the proximal vessel.
ostia. This angiogram demonstrates mild to moderate narrowing of
the left renal artery.
with a 108 to 208 LAO angulation. Occasionally, additional

slight (108) cranial or caudal angulation may be necessary for
catheter can also be used. In cases of aortoiliac tortuosity, a optimal visualization. Since the right renal artery travels down
Cobra C2 (Terumo, Somerset, New Jersey, U.S.), Sos, or Sim- below the inferior vena cava, additional RAO projections may
mons catheter may be necessary to reach the ostia (Fig. 24.13). be necessary to adequately visualize the entire course of this
Regardless of the equipment chosen, the catheter must never be vessel. Care should be taken to visualize any accessory renal
advanced to the renal ostia without the use of a wire. Such arteries since these occur commonly.
manipulation can result in advancement of atheromatous Prior to injection, the catheter must be aspirated well and
debris into the renal ostium (“snowplow effect”). Using a the pressure waveform should reflect a crisp arterial tracing.
0.035-in. wire, the catheter should be positioned anteriorly at The field of view should be large enough to incorporate the
the L1 level. Gentle retraction with clockwise (left renal artery) kidney, and the cineangiogram run should also be long enough
or counterclockwise (right renal artery) manipulation of the to visualize the influx of contrast into the renal cortex. This
catheter allows for selective engagement. nephrogram yields important insight into renal size and
Aortic calcifications and prior-placed renal artery stents regional function. Such adjunctive information becomes partic-
serve as important landmarks indicating the renal ostia ularly important in individuals with suboptimal or equivocal
(Fig. 24.14). Bear in mind, the renal arteries do not originate noninvasive studies.
directly from the lateral aorta. The right renal artery originates With selective renal angiography, a few additional techni-
about 208 to 308 anteriorly, and the left originates only slightly cal points should be mentioned. The operator should be careful
posteriorly. Therefore, PA projection angiography may not to avoid unnecessary trauma to the renal ostia with engagement
adequately illustrate the renal ostium. Accordingly, one study as this may lead to showering of aorto-ostial atherosclerotic
demonstrated optimal visualization of the right renal ostia in debris. Care should also be taken not to deep-seat the catheter
only 26% of cases and of the left in 38% of cases (32). Our past the renal ostium as this may lead to underestimation of the
practice is to perform selective right and left renal angiography lesion severity due to insufficient opacification of the ostium.
[2/7/010/8:56:54] [297–306]
Damping of the pressure waveform on selective engagement CONCLUSIONS

may indicate significant ostial stenosis. The invasive assessment of patients with suspected mesenteric
or renal vascular disease needs to be performed cautiously and
meticulously. These patients, particularly those with preexist-
SPECIAL ISSUES/CONSIDERATIONS ing renal insufficiency, can have a high risk of procedure-
Contrast Selection and Renal Insufficiency related complications. However, if careful technique is
An optimal diagnostic angiographic should allow for complete observed and minimal contrast used, such angiographic assess-
examination of the arterial bed with the least amount of con- ment can yield critical clinical information and potentially set
trast used. In patients with renal insufficiency, it is critical to the stage for important interventional therapies.
minimize the amount of contrast used during the angiographic
study. Low osmolar contrast agents are preferred as they min-
imize patient discomfort and both reduce the risk of contrast- REFERENCES
induced nephropathy (CIN) and allergic reactions. The only 1. Hirsch AT, Haskal ZJ, Hertzer NR, et al. ACC/AHA 2005 Practice
available iso-osmolal contrast agent, iodixinol, may be associ- guidelines for the management of patients with peripheral arterial
ated with a lower risk of nephropathy than low osmolal agents, disease (lower extremity, renal, mesenteric, and abdominal aortic):
particularly among patients at high risk for contrast nephrop- a collaborative report from the American Association for Vascular
Surgery/Society for Vascular Surgery, Society for Cardiovascular
athy (35). Despite the lower risk of complications with these
Angiography and Interventions, Society for Vascular Medicine
agents, the possibility of CIN and allergic reaction remain. and Biology, Society of Interventional Radiology, and the ACC/
Carbon dioxide (CO2) angiographic imaging is an alternative AHA Task Force on Practice Guidelines (Writing Committee to
in patients who remain at high risk for these complications. Develop Guidelines for the Management of Patients With Periph-
CO2 is a very dissolvable gas that is nontoxic to the kidneys and eral Arterial Disease): endorsed by the American Association of
is nonallergic (36,37). It works by displacing blood cells inside Cardiovascular and Pulmonary Rehabilitation; National Heart,
the vessel, effectively reducing the radiographic density within Lung, and Blood Institute; Society for Vascular Nursing; Trans-
the lumen. There are certain limitations to this form of imaging Atlantic Inter-Society Consensus; and Vascular Disease Founda-
that should be noted. Importantly, contrast resolution with CO2 tion. Circulation 2006; 113:e463–e654.
angiography, while often adequate, is inferior to that with 2. McKinsey JF, Gewertz BL. Acute mesenteric ischemia. Surg Clin
North Am 1997; 77:307–318.
traditional iodinated contrast agents (38). CO2 angiography
3. Bradbury AW, Brittenden J, McBride K, et al. Mesenteric ischae-
must also be performed in a controlled fashion with carefully mia: a multidisciplinary approach. Br J Surg 1995; 82:1446–1459.
administered gas volumes to avoid gas trapping in the pulmo- 4. Boley SJ, Brandt LJ, Sammartano RJ. History of mesenteric ische-
nary circulation (39). Gadolinium also has been studied as a mia. The evolution of a diagnosis and management. Surg Clin
possible alternative agent in patients at risk for CIN. However, North Am 1997; 77:275–288.
this agent has not been proven to have any significant advan- 5. Cappell MS. Intestinal (mesenteric) vasculopathy. I. Acute supe-
tage over low osmolar agents in reducing CIN and has been rior mesenteric arteriopathy and venopathy. Gastroenterol Clin
largely abandoned (40). North Am 1998; 27:783–825, vi.
6. Reinus JF, Brandt LJ, Boley SJ. Ischemic diseases of the bowel.
Gastroenterol Clin North Am 1990; 19:319–343.
Difficulty in Visualizing Renal Ostia 7. Brandt LJ, Boley SJ. AGA technical review on intestinal ischemia.
Renal artery ostial disease may be missed by standard angiog- American Gastrointestinal Association. Gastroenterology 2000;
raphy. If clinical suspicion of RAS is high, nontraditional views 118:954–968.
with added cranial or caudal angulation should be attained. 8. Wilcox MG, Howard TJ, Plaskon LA, et al. Current theories of
Intravascular ultrasound (IVUS) is an alternative modality that pathogenesis and treatment of nonocclusive mesenteric ischemia.
Dig Dis Sci 1995; 40:709–716.
can be used for better visualization of the renal arteries and to
9. Fisher DF Jr., Fry WJ. Collateral mesenteric circulation. Surg
determine the presence of renal artery ostial disease. The Gynecol Obstet 1987; 164:487–492.
operator should pay attention to hemodynamic clues such as 10. Buchardt Hansen HJ. Abdominal angina: results of arterial recon-
catheter damping that may indicate a significant ostial lesion. struction in 12 patients. Acta Chir Scand 1976; 142:319–325.
Finally, measurement of fractional flow reserve across interme- 11. Mikkelsen W. Intestinal angina: its surgical significance. Surg
diate lesions has emerged as an attractive technique to assess Gynecol Obstet 1957; 94:262–267.
indeterminate lesions. A translesional pressure gradient of 12. Korotinski S, Katz A, Malnick SD. Chronic ischaemic bowel dis-
20 mmHg or greater is generally considered indicative of a eases in the aged—go with the flow. Age Ageing 2005; 34:10–16.
hemodynamically significant stenosis. 13. Dunbar JD, Molnar W, Beman FF, et al. Compression of the celiac
trunk and abdominal angina. Am J Roentgenol Radium Ther Nucl
Med 1965; 95:731–744.
Difficulty Using Arm Approach 14. Bech F, Loesberg A, Rosenblum J, et al. Median arcuate ligament
As discussed above, an arm approach for mesenteric and renal compression syndrome in monozygotic twins. J Vasc Surg 1994;
angiography can be very helpful in certain circumstances. 19:934–938.
However, in cases of subclavian stenosis, innominate stenosis, 15. Holland AJ, Ibach EG. Long-term review of coeliac axis compres-
sion syndrome. Ann R Coll Surg Engl 1996; 78:470–472.
or tortuous aortic arch, the passage of diagnostic catheters and
16. Horton KM, Talamini MA, Fishman EK. Median arcuate ligament
equipment from an arm approach can be challenging. In these syndrome: evaluation with CT angiography. Radiographics 2005;
cases, a long sheath should be used and advanced into the 25:1177–1182.
descending aorta and placed above the level of the mesenteric 17. Karahan OI, Kahriman G, Yikilmaz A, et al. Celiac artery com-
vessels. For right radial approach, catheters that are 125 cm in pression syndrome: diagnosis with multislice CT. Diagn Interv
length must be used. A stiff 0.035-in. wire such as a Rosen or a Radiol 2007; 13:90–93.
stiff Glidewire may also facilitate catheter and sheath place- 18. Messina LM, Shanley CJ. Visceral artery aneurysms. Surg Clin
ment in this circumstance. North Am 1997; 77:425–442.
[2/7/010/8:56:54] [297–306]
19. Shanley CJ, Shah NL, Messina LM. Uncommon splanchnic artery 31. Perdu J, Boutouyrie P, Bourgain C, et al. Inheritance of arterial
aneurysms: pancreaticoduodenal, gastroduodenal, superior mesen- lesions in renal fibromuscular dysplasia. J Hum Hypertens 2007;
teric, inferior mesenteric, and colic. Ann Vasc Surg 1996; 10:506–515. 21:393–400.
20. Hallett JW Jr. Splenic artery aneurysms. Semin Vasc Surg 1995; 8: 32. Rajaopalan S, Mukherjee D, Mohler E, eds. Manual of Vascular
321–326. Diseases. Philadelphia: Lippincott Williams & Wilkins, 2005:
21. Kasirajan K, Greenberg RK, Clair D, et al. Endovascular manage- 202–204.
ment of visceral artery aneurysm. J Endovasc Ther 2001; 8:150–155. 33. Beregi JP, Mauroy B, Willoteaux S, et al. Anatomic variation in the
22. Nosher JL, Chung J, Brevetti LS, et al. Visceral and renal artery origin of the main renal arteries: spiral CTA evaluation. Eur
aneurysms: a pictorial essay on endovascular therapy. Radio- Radiol 1999; 9:1330–1334.
graphics 2006; 26:1687–1704; quiz 34. Kadir S. Angiography of the Kidneys. Philadelphia: Saunders,
23. Henke PK, Cardneau JD, Welling TH III, et al. Renal artery 1986.
aneurysms: a 35-year clinical experience with 252 aneurysms in 35. Aspelin P, Aubry P, Fransson SG, et al. Nephrotoxic effects in
168 patients. Ann Surg 2001; 234:454–462; discussion 62–63. high-risk patients undergoing angiography. N Engl J Med 2003;
24. Scoble J. Renal Vascular Disease. London: WB Saunders Co Ltd, 348:491–499.
1996. 36. Shaw DR, Kessel DO. The current status of the use of carbon
25. Conlon PJ, Little MA, Pieper K, et al. Severity of renal vascular dioxide in diagnostic and interventional angiographic procedures.
disease predicts mortality in patients undergoing coronary Cardiovasc Intervent Radiol 2006; 29:323–331.
angiography. Kidney Int 2001; 60:1490–1497. 37. Liss P, Eklof H, Hellberg O, et al. Renal effects of CO2 and
26. Rihal CS, Textor SC, Breen JF, et al. Incidental renal artery stenosis iodinated contrast media in patients undergoing renovascular
among a prospective cohort of hypertensive patients undergoing intervention: a prospective, randomized study. J Vasc Interv
coronary angiography. Mayo Clin Proc 2002; 77:309–316. Radiol 2005; 16:57–65.
27. Leertouwer TC, Gussenhoven EJ, Bosch JL, et al. Stent placement 38. Schreier DZ, Weaver FA, Frankhouse J, et al. A prospective study
for renal arterial stenosis: where do we stand? A meta-analysis. of carbon dioxide-digital subtraction vs standard contrast arte-
Radiology 2000; 216:78–85. riography in the evaluation of the renal arteries. Arch Surg 1996;
28. Novick AC, Ziegelbaum M, Vidt DG, et al. Trends in surgical 131:503–507; discussion 7–8.
revascularization for renal artery disease. Ten years’ experience. 39. Kessel DO, Robertson I, Patel JT, et al. Carbon-dioxide-guided
JAMA 1987; 257:498–501. vascular interventions: technique and pitfalls. Cardiovasc Inter-
29. Safian RD, Textor SC. Renal-artery stenosis. N Engl J Med 2001; vent Radiol 2002; 25:476–483.
344:431–442. 40. Boyden TF, Gurm HS. Does gadolinium-based angiography pro-
30. Schreiber MJ, Pohl MA, Novick AC. The natural history of tect against contrast-induced nephropathy?: a systematic review
atherosclerotic and fibrous renal artery disease. Urol Clin North of the literature. Catheter Cardiovasc Interv 2008; 71:687–693.
Am 1984; 11:383–392.
[21/6/010/21:0:43] [307–313]
25
Peripheral arterial angiography

Ivan P. Casserly
INTRODUCTION normal circumstances, it provides no significant angiographic

The widespread availability of noninvasive angiographic tech- branches in the thigh. While in the thigh, it sequentially runs
niques, notably computed tomography (CT) and magnetic res- through the femoral triangle, the muscular adductor canal, and
onance (MR) angiography, has had a dramatic impact on the the adductor hiatus, at which point it becomes the popliteal
practice of routine peripheral angiography of the upper and artery. The popliteal artery initially runs in the popliteal fossa
lower extremities (1–3). Most invasive angiographic studies are posterior to the distal third of the femur, and subsequently
now performed to confirm the findings on noninvasive studies along the posterior surface of the tibial plateau, providing
for the purpose of performing vascular intervention or to image articular branches to the capsule and ligaments of the knee
small diameter extremity vessels for which the resolution of CT joint, and sural branches that supply the calf muscles (Fig. 25.2).
or MR angiography is insufficient. The result is a more targeted Although the popliteal artery is typically described as having a
and individualized approach to angiography that answers a trifurcation, this is a misnomer, as the most typical pattern is a
clinical question or helps direct a therapy. This chapter will bifurcation into an anterior tibial (AT) artery and tibioperoneal
outline the angiographic anatomy of the arterial supply of the trunk (Fig. 25.4). Passing anteriorly through the interosseous
upper and lower extremities, and provide a practical approach membrane between the tibia and fibula, the AT reaches the
to angiography of these territories. anterior compartment of the leg and courses inferiorly to a
point midway between the malleoli where it becomes the
dorsalis pedis artery. The tibioperoneal trunk has a variable
LOWER EXTREMITY length (typically 2–4 cm), before dividing into the posterior
Anatomic Considerations tibial (PT) and peroneal arteries. Both of these arteries continue
Arterial Anatomy of the Lower Extremities inferiorly in the posterior compartment of the leg. The PT
The distal abdominal aorta typically bifurcates at the level of passes posterior to the medial malleolus before dividing into
third or fourth lumbar vertebra into right and left common iliac medial and lateral plantar branches (Figs. 25.5 and 25.6),
arteries (CIA) (Fig. 25.1) (4,5). Each CIA passes in an anterior whereas the peroneal branch typically terminates above the
direction in the pelvis before bifurcating into external (EIA) and level of the ankle joint. All three tibial branches provide the major
internal (IIA) iliac artery branches. The IIA generally divides arterial supply to the calf muscles. Although variable, the termi-
into two trunks (in 60% of patients) but may have four or more nal portions of the three tibial branches typically provide impor-
trunks. In addition to branches to the pelvic organs (i.e., blad- tant collateral communications that provide flow to the foot in the
der, rectum, female, and male reproductive organs), the IIA presence of proximal occlusive disease of one or more tibial
provides the superior and inferior gluteal branches that supply branches (Fig. 25.7).
the gluteal muscles and the obturator branch. These branches Within the foot, the plantar branches of the PT, and the
serve as an important source of collaterals to the ipsilateral and dorsalis pedis branch of the AT, variably contribute arterial
contralateral limb in the presence of iliac occlusive disease. inflow to the metatarsal arch, which runs along the base of the
The EIA continues with an anterior trajectory, becoming metatarsal bones. The metatarsal branches arise from this arch
the common femoral artery (CFA) at the level of the inguinal and course toward the interdigital spaces, providing digital
ligament. There is some variability in the location of the CFA branches to the toes.
bifurcation into superficial (SFA) and profunda (PFA) branches
(Fig. 25.2). One prospective angiographic study found the Arterial Anatomy of the Upper Extremities
bifurcation below the middle and inferior border of the femoral The arterial inflow to the right and left upper extremities is
head in 99% and 80% of individuals, respectively (6). Arising provided by the right and left subclavian arteries (SCA), respec-
from the proximal portion of the PFA are the medial and lateral tively (3,7). While the left subclavian artery arises directly from
circumflex femoral branches, which serve as important collat- the aortic arch as the third and terminal great vessel, the right
eral pathways in the presence of occlusive disease affecting subclavian arises from the bifurcation of the innominate artery.
the iliac arteries and CFA. The PFA continues laterally into In approximately 0.5% of patients, the right subclavian artery
the thigh supplying three or four perforator branches to the arises as the terminal vessel from the descending thoracic aorta
muscles of the thigh, which serve as the main source of (Fig. 25.8).
collaterals to the leg when the SFA is occluded (Fig. 25.3), The subclavian artery extends from its origin to the
underscoring the importance of the PFA. Fortunately, the PFA lateral border of the first rib (Fig. 25.9). Its course is divided
rarely has significant occlusive disease in patients with exten- into three segments based on the relationship of the artery to
sive disease in the SFA. the scalenus anterior muscle. The first segment that lies medial
The primary function of the SFA is to provide an arterial to the scalenus anterior muscle provides the most important
conduit for arterial flow to the knee, leg, and foot. Under branches of the subclavian artery, notably the vertebral and
[21/6/010/21:0:43] [307–313]
Figure 25.1 Angiographic anatomy of the distal abdomi-

nal aorta and pelvic arteries. (A) Right anterior oblique
view; (B) left anterior oblique view. 1, Inferior mesenteric
artery; 2, lumbar branch; 3 and 4, right and left common
iliac artery; 5 and 6, right and left internal iliac artery;
7 and 8, right and left external iliac artery; 9 and 10, right
and left common femoral artery.
Figure 25.2 Angiographic anatomy of the right common femoral

(A), profunda femoral (A), superficial femoral (B), popliteal (B,C),
and tibial arteries (C). 1, Common femoral artery; 2, profunda
femoral artery (PFA); 3, superfical femoral artery (SFA); 4, perforat-
ing branch of PFA; 5,6, muscular branches of SFA; 7, distal SFA; 8,
approximate location of adductor hiatus; 9, descending genicular Figure 25.3 (A) Collaterals branches from right superficial femoral
branch; 10, popliteal artery; 11, articular branch of popliteal artery; artery (SFA) in presence of significant stenosis (arrowhead); (B)
12, sural branch of popliteal artery; 13, anterior tibial artery; 14, collateral filling of distal right SFA and popliteal artery via collaterals
tibioperoneal trunk; 15, peroneal artery; 16, posterior tibial artery. from the profunda femoral artery (PFA) in the presence of a distal
SFA occlusion (extent of occlusion shown by arrows). 1, SFA; 2,
SFA collateral; 3, PFA; 4, PFA collateral to distal SFA/popliteal
artery; 5, popliteal artery.
internal thoracic arteries and the thyrocervical trunk. While the

vertebral and internal thoracic arteries are largely constant in
their origin and course, the thyrocervical trunk greatly varies raphy, the latter anatomic border roughly corresponds to the
between individuals, in terms of the pattern and size of its position of the anatomical neck of the humerus. Similar to the
various branches. In 4% to 6% of individuals, the left vertebral subclavian artery, the axillary artery is divided into three
artery had a direct origin from the aortic arch, typically segments on the basis of its relationship to the pectoralis
between the origins of the left common carotid and left sub- minor muscle. Significant interindividual variation is seen in
clavian arteries. the location and size of branches of the axillary artery.
The axillary artery reflects the continuation of the sub- The brachial artery is the anatomic equivalent to the SFA
clavian artery and extends from the lateral border of the first rib of the lower extremity. As the direct continuation of the axillary
to the inferior border of the teres major muscle. During angiog- artery, it runs through the arm to the neck of the radius, where
[21/6/010/21:0:43] [307–313]
PERIPHERAL ARTERIAL ANGIOGRAPHY 309
Figure 25.4 Examples of variant angiographic anatomy of the tibial

vessels. (A) True trifurcation of the left popliteal artery; (B) long
tibioperoneal trunk in right leg; (C) bifurcation of the left peroneal
artery; (D) absent tibioperoneal trunk, with origin of the posterior
tibial artery as the first branch of the left popliteal artery, followed by Figure 25.6 Angiographic anatomy of the left foot. (A) Lateral
the peroneal artery and anterior tibial artery. 1, Popliteal artery; 2, projection; (B) posterior-anterior projection with cranial angulation.
anterior tibial artery; 3, tibioperoneal trunk; 4, peroneal artery; 5, 1, Anterior tibial artery; 2, peroneal artery; 3, posterior tibial artery;
posterior tibial artery; 6 and 7, bifurcation branches of peroneal 4, dorsalis pedis artery; 5, lateral plantar artery; 6, medial plantar
artery—one of these branches supplied the medial plantar branch to artery; 7, metatarsal arch; 8, metatarsal branches.
the foot.
Figure 25.5 Angiograms from two patients (A,B) demonstrating

the collateral circulation between the distal portion of all three tibial Figure 25.7 Angiographic anatomy of the right foot in a patient with
vessels. 1, Anterior tibial artery; 2, peroneal artery; 3, posterior tibial an occluded dorsalis pedis artery. (A) Lateral projection; (B) posterior-
artery; 4, posterior communicating branch; 5, perforating branch; 6, anterior projection with cranial angulation. 1, posterior tibial artery;
dorsalis pedis artery; 7, lateral plantar artery; 8, medial plantar 2, peroneal artery; 3, lateral plantar artery; 4, medial plantar artery;
artery; 9, metatarsal arch. 5, metatarsal arch; 6, metatarsal branches.
[21/6/010/21:0:43] [307–313]
Figure 25.8 Anomalous origin of the right subclavian artery from the arch of the aorta distal to the origin of the left subclavian artery (arteria
lusoria). (A) LAO projection; (B) RAO projection. 1, right common carotid artery; 2, left common carotid artery; 3, left subclavian artery; 4, right
subclavian artery.
Figure 25.10 Angiographic anatomy of the brachial artery.

1, Brachial artery; 2, profunda brachi artery; 3, radial artery;
4, interosseous artery; 5, ulnar artery; 6, recurrent radial branch;
7, recurrent ulnar branch.
Figure 25.9 Angiographic anatomy of the right subclavian artery.

1, Subclavian artery; 2, vertebral artery; 3, right internal mammary
artery; 4, thryocervical trunk; 5, suprascapular branch; 6, axillary
artery; 7, circumflex humeral branch. Arrow indicates presence of and is usually larger than the radial artery. It arises from the
critical heavily calcified stenosis at junction of right subclavian with brachial artery bifurcation at the level of the neck of the radius
innominate artery. and runs through the arm to the pisiform carpal bone. Its main
branch is the interosseous artery, which courses lateral to the
ulnar artery, supplying the forearm muscles and interosseous
membrane. In addition, the ulnar artery supplies branches to
it divides into radial and ulnar branches (Fig. 25.10). The the elbow joint, muscular branches to the forearm, carpal
profunda brachi branch of the brachial artery is considerably branches to the palmar and dorsal aspect of the wrist, contrib-
smaller and clinically insignificant compared with its counter- utes to the deep palmar arch, and continues into the hand as the
part in the lower extremity (i.e., the PFA). Other major branches major source of blood flow to the superficial palmar arch.
of the brachial artery include muscular branches to the arm Analogous to the AT artery of the lower extremity, the
muscles, the nutrient artery to the humerus, and vessels to the radial artery is the smaller of the terminal branches of the
elbow joint. brachial artery. It runs from the brachial bifurcation toward
Analogous to the posterior tibial artery of the lower the styloid process of the radius, contributing flow to the elbow
extremity, the ulnar artery is the major vessel to the forearm joint, muscular branches to the forearm, and carpal branches to
[21/6/010/21:0:43] [307–313]
potentially nephrotoxic iodinated contrast agents, it is impor-

tant to incorporate data from the clinical history, physical exam
(i.e., pulse exam), and noninvasive hemodynamic studies to
help target the angiographic procedure.
The use of the iso-osmolar contrast agent iodixanol (Vis-
ipaque, GE Healthcare) has dramatically reduced the discomfort
experienced by patients during peripheral angiography, partic-
ularly when imaging the distal vessels, and is strongly recom-
mended. Recently, the availability of automated injection
systems (e.g., MEDRAD Avanta Injection System, ACIST CVi
contrast delivery system) has greatly simplified the administra-
tion of contrast for peripheral angiography and offers improved
safety over the use of older manifold systems (10). Finally, it is
important that peripheral angiography be performed using X-
ray equipment that is equipped to provide for optimal vascular
imaging, with the ability to subtract nonvascular structures
(digital subtraction angiography, DSA). These systems have
additional features that simplify and improve the safety of
vascular angiography (e.g., road mapping, bolus chase capabil-
ity, 3-D angiography). Performing peripheral angiography with
X-ray equipment without dedicated vascular imaging capability
Figure 25.11 Angiographic anatomy of the left hand in a patient is strongly discouraged. Tables 25.1 and 25.2 summarize the
with vasculitis and occlusion of the ulnar artery (black arrow). most commonly used catheters and wires used during lower
1, Radial artery; 2, ulnar artery; 3, interosseous artery; 4, deep and upper extremity angiography.
plantar arch; 5, superficial plantar arch; 6, common palmar digital
branch; 7, proper palmar digital branch; 8, digital pad network. Lower Extremity Angiography
Lower extremity angiography generally encompasses imaging
the distal abdominal aorta and pelvic vessels, in addition to the
lower extremity vessels. This is most commonly achieved using
the palmar and dorsal aspects of the wrist. In the hand, it retrograde femoral access. Using this access, a flush catheter is
contributes to the superficial palmar arch and continues into
the hand as the major source of flow to the deep palmar arch. Table 25.1 List of Catheters Used During Upper and Lower
The superficial and deep palmar arches provide the Extremity Angiography and Their Function
arterial flow to the hand (Fig. 25.11). Angiographically, the
Function
superficial arch lies distal to the deep arch and is generally
more prominent. Formed primarily from the terminal portion Contralateral Subclavian
of the ulnar artery, the superficial arch is less commonly com- Catheter iliac access access Angiography
plete (80%). The deep arch is formed primarily from the termi- Internal þ þ
nal portion of the radial artery and is complete in the majority mammary
of individuals (*95%). Common palmar digital arteries arise SoS þ þ
from the superficial arch and fuse with palmar metacarpal Simmons þ þ þ
branches from the deep arch. In the interdigital space, each of Vitek þ þ
JR 4 þ þ
the common palmar digital arteries divides into two proper
Bernstein þ þ
palmar digital arteries that run along the borders of the 2nd to Glide catheter þ þ
5th digits and supply the arterial network of the finger pads. Flush catheter þ
The princes pollicis branch typically arises from the terminal
portion of the radial artery and supplies the thumb.
Table 25.2 List of Wires Typically Used During Upper and Lower
Extremity Angiography and Their Function
PERIPHERAL ANGIOGRAPHIC TECHNIQUE
General Comments Function
As outlined earlier, the current approach to peripheral angiog- Routine Provide Negotiate
raphy is typically targeted to confirm the findings of a prior Wire use support tortuosity
noninvasive imaging study and guide an endovascular inter- J-tipped wire þ
vention or to define anatomic detail that cannot be provided by Wholey wire þ þ
such imaging modalities, such as the tibial and pedal anatomy Magic Torque þ þ
in a patient with critical limb ischemia. With the recent reports Supra-Core þ
of nephrogenic systemic fibrosis following the administration Amplatz
of gadolinium-based contrast agents with MR angiography in Extra-stiff þ
patients with renal insufficiency (8,9), there has been a minor Super-stiff þ
resurgence in demand for routine contrast angiography to Glidewire
anatomically assess the peripheral arterial anatomy. However, Floppy þ
Stiff þ þ
even in such circumstances, in an effort to limit exposure to
[21/6/010/21:0:43] [307–313]
advanced over a wire to the level of approximately L3 in the provide views without vessel overlap. High-quality imaging of
distal abdominal aorta. With the catheter in this position, left the foot is generally reserved in the assessment of patients with
anterior oblique (LAO) 308 to 408 and right anterior oblique critical limb ischemia to assess the dominant arterial inflow to
(RAO) 308 to 408 pelvic angiograms are acquired with the the foot or the site of tissue loss. In the authors’ experience, the
patient holding their breath (typically in exhalation). Injection combination of a steep contralateral oblique image of the
rates of 10 to 15 cc/sec for a total of 20 to 30 cc are adequate in externally rotated foot and a cranial image of the foot in the
most patients (using psi of 800–1000). Acquisition of these neutral position allows the best assessment of the small vessel
oblique angulations is important to allow for optimal visual- anatomy of the foot for clinical decision making (Figs. 25.5 and
ization of eccentric lesions, and definition of obstructive disease 25.6). Peripheral X-ray angiographic imaging systems allow a
involving the aortoiliac and common iliac bifurcations programmed time delay between the injection of contrast and
(e.g., LAO to view the right aortoiliac and common iliac image acquisition. This becomes increasingly important with
bifurcations). prolonged transit times (i.e., in the presence of total occlusions)
Below the level of the iliac arteries, angiography of the and increased distance between the site of contrast injection
lower extremity may be performed using one of two basic and image acquisition (i.e., the foot with the catheter placed in
methods: the bolus chase technique or sequential static imag- the CFA).
ing. Using specially equipped angiographic X-ray systems, the To imaging the contralateral limb using the sequential
bolus chase technique is achieved by administering a large static technique, a diagnostic catheter must be placed in the
volume of contrast in the lower abdominal aorta to image both contralateral EIA or CFA. This clearly requires patency of the
lower extremities (*30 cc/sec for a total of 90 cc) or the EIA to iliac system, and the major factors impacting the complexity of
image an individual lower extremity (*15 cc/sec for a total of this task include the angulation of the aortic bifurcation and the
45 cc) that is coordinated with activation of cineangiographic tortuosity of the iliac system. In most patients, an internal
acquisition. Depending on the particular X-ray system, the table mammary (IM) artery catheter allows for successful engage-
is subsequently moved proximally using an automated or ment of the contralateral CIA. However, for patients with acute
manual (according to the speed of the contrast runoff) mecha- angulation of the aortic bifurcation, prior aortoiliac stent place-
nism, allowing imaging of the entire lower extremity. While ment or prior aortobifemoral bypass surgery, an SoS (Angio-
this method certainly decreases the total procedure time and Dynamics, Queensbury, New York, U.S.) or Simmons (Cordis
X-ray dose required to obtain angiographic images of the Endovascular) catheter may be required. Following successful
extremity, and eliminates the need to selectively engage placement of a diagnostic catheter in the contralateral CIA, a
the individual limb vessels, it has a number of limitations. In road map of the iliac system (using 3 cc/sec for total of 6 cc) is
the authors’ experience, the definition of the infrapopliteal performed with the image intensifier in the contralateral obli-
arterial anatomy using this technique is typically poor, and que projection. This allows passage of a stiff angled Glidewire
the presence of asymmetric occlusive disease, which creates (Terumo, Somerset, New Jersey, U.S.) into the contralateral
significant differences in the rate of contrast runoff in both CFA. Severe angulation of the aortic bifurcation or iliac tor-
lower extremities, invariably results in suboptimal imaging tuosity may require that the wire be advanced further into the
from one of the limbs. The inability to alter the angulation of PFA or SFA to provide increased wire purchase and support.
image acquisition along the length of the lower extremity also Typically, the IM catheter is then advanced over this wire,
decreases the sensitivity of the technique for the detection of which is exchanged for a Super stiff Amplatz wire (with 1-cm
disease, particularly at bifurcation points. As a result, while the soft tip) (Boston Scientific Corp., One Boston Scientific Place,
bolus chase technique is promoted as a method to limit contrast Massachusetts, U.S.) or Supra-Core wire (Abbott Vascular,
use, if additional static imaging is required to answer remain- Santa Clara, California, U.S.) for support. The IM catheter is
ing diagnostic dilemmas, overall contrast use may be increased. then exchanged for a straight flush catheter through which
Although more time consuming and requires more cath- angiography of the extremity is performed. Occasionally, the
eter manipulation, sequential static imaging of the lower IM catheter cannot be advanced over the stiff-angled glidewire.
extremity remains the authors’ preferred method for lower In this case, a more flexible catheter such as a Glide catheter
extremity angiography. Image quality is usually superior, and (Terumo) may be used, and by adopting a strategy of using
with careful technique, the total contrast use is often compara- increasingly supportive wires and catheters, the straight flush
ble to that using the bolus chase method. Imaging of the lower catheter can be ultimately delivered, although angiography
extremity ipsilateral to the femoral sheath is achieved by through a glide catheter is also possible.
injecting contrast through the sidearm of the sheath, or through The principles for imaging the contralateral limb are
a flush catheter positioned in the ipsilateral EIA. By moving identical to those for the ipsilateral limb, as outlined earlier.
either the image intensifier or the X-ray table, the entire region In selected patient subsets, such as those with renal insuffi-
of interest is sequentially imaged. With modern image intensi- ciency, and/or patients with critical limb ischemia requiring
fiers with a 20 in. field of view, the lower extremity may be high-quality images of the infrapopliteal anatomy, the flush
imaged using three to four overlapping acquisitions. Injection catheter may be advanced more distally over a wire into the
rates of 3 to 5 cc/sec for a total of 9 to 15 cc (using psi of distal SFA or popliteal artery with the goal(s) of reducing
500–600) will provide uniform opacification of the entire length contrast use and improving image quality.
of vessel within the field of view.
An ipsilateral oblique projection (*308 is optimal for Alternative Arterial Access for Lower Extremity Angiography
imaging the CFA and its bifurcation into the PFA and SFA. Both antegrade CFA access and upper extremity arterial access
Between the level of the mid-SFA and ankle, images are gen- (brachial or radial artery) may also be used to perform lower
erally acquired in the posteroanterior (PA) or ipsilateral oblique extremity angiography. While antegrade CFA access is typi-
projection (15–258). In the leg, the relationship of the tibial cally used for complex lower extremity interventional proce-
vessels is variable, and multiple views may be required to dures, it may also be necessary for diagnostic purposes where
[21/6/010/21:0:43] [307–313]
the iliac anatomy prohibits contralateral access. Upper extrem- be exchanged for a gentle shaped catheter (e.g., angled glide,
ity arterial access may be required in patients who have a straight flush, Bernstein) that can be delivered to the axillary
contraindication to CT or MR angiography and have complex artery after the 0.035 in. wire has been advanced into the brachial
iliac disease or an aortic occlusion preventing retrograde CFA artery.
access. The angiographic principles of lower extremity angiog- Axillary artery angiography is performed with the arm in
raphy from these alternative access sites are identical to that the neutral position or slightly abducted. Angiography of the
described earlier. arm, forearm, and hand is performed with the patients forearm
and hand placed in the supine position on an arm board, and
the digits splayed apart. Tape is required to maintain the hand
Upper Extremity Angiography in this position and avoid motion artifact. Prior to imaging of
Diagnostic angiography of the upper extremities is typically the hand vessels, the administration of vasodilators is recom-
performed using retrograde femoral access. An initial assess- mended, and the surface temperature of the hand should be
ment of the anatomy of the aortic arch is helpful in determining kept warm to remove any component of spasm. After achieving
the appropriate diagnostic catheter for engagement of the right the appropriate positioning of the patients upper extremity,
and left SCA and the detection of anomalies (e.g., anomalous sequential overlapping static images of the upper extremity
origin of right SCA distal to left SCA, direct origin of the vessels are acquired in the PA projection, using angulated
vertebral artery from the arch) that might influence these views when required. Again, the use of an automated power
engagements. In the authors’ experience, good quality angiog- injector that allows the injection of 12 cc over a 3-second period
raphy of the aortic arch is achieved using a 6-Fr pigtail catheter (psi of 500 to 600) for a 20 in. field of view is optimal for
placed in the ascending thoracic aorta, and injecting 40 cc of imaging above the level of the hand. In the hand, higher
contrast at a rate of 20 cc/sec and a pressure limit of 1000 psi magnification images are required with use of shorter injection
with the image intensifier or flat panel in the LAO 408 position. lengths (e.g., 5 cc/sec for total of 10 cc).
In patients with normal anatomic origin of the great vessels
and a benign configuration of the arch (i.e., type I), the innomi-
CONCLUSIONS
nate artery and left subclavian artery can be selectively engaged
Peripheral angiography of the upper and lower extremities
with either a Bernstein (Cordis Endovascular, Miami Lakes,
remains important in the assessment of patients with a variety
Florida, U.S.), Judkins right (JR) 4, or angled glide diagnostic
of vascular disorders. A thorough understanding of the normal
catheter. With increasing complexity of the aortic arch (i.e., type
and anomalous arterial anatomy of the upper and lower extrem-
III), a Vitek catheter (Cook Inc., Bloomington, Indiana) or Sim-
ities and a thoughtful approach to the angiographic technique
mons catheter (Cordis Endovascular) may be required. While the
will maximize the angiographic information required to answer
Vitek catheter can typically be shaped in the descending thoracic
specific clinical questions or guide endovascular intervention.
aorta, the Simmons catheters (i.e., Simmons 1, 2, and 3) require
more skilled manipulation in the aortic arch and great vessels to
help shape the catheter for vessel engagement. REFERENCES
The bifurcation of the innominate artery (i.e., including 1. Ersoy H, Rybicki FJ. MR angiography of the lower extremities.
the ostium and proximal portion of the right SCA) is best AJR Am J Roentgenol 2008; 190:1675–1684.
2. Schernthaner R, Stadler A, Lomoschitz F, et al. Multidetector CT
imaged in the RAO oblique projection, whereas the LAO pro-
angiography in the assessment of peripheral arterial occlusive
jection typically provides the best assessment of the origins of disease: accuracy in detecting the severity, number, and length of
the right vertebral and right internal mammary arteries. For the stenoses. Eur Radiol 2008; 18:665–671.
left SCA, the RAO projection generally allows the most accurate 3. Stepansky F, Hecht EM, Rivera R, et al. Dynamic MR angiography
assessment of the origins of the left vertebral artery and left of upper extremity vascular disease: pictorial review. Radio-
internal mammary artery (LIMA). When imaging the SCA graphics 2008; 28:e28.
specifically for the purpose of assessing the presence of arterial 4. Kadir S. Abdominal aorta and pelvis. In: Atlas of Normal and
compression due to thoracic outlet syndrome, angiography Variant Angiographic Anatomy. 1st ed. Philadelphia: W.B. Saun-
should be performed in the PA projection with the arm in the ders, 1991:97–125.
neutral position (i.e., arm at side in adducted position) and 5. Kadir S. Arterial anatomy of the lower extremities. In: Kadir S, ed.
Normal and Variant Angiographic Anatomy. 1st ed. Philadelphia:
repeated with the shoulder in full abduction, external rotation,
W.B. Saunders, 1991:123–160.
and retroversion (as if pitching a baseball). 6. Garrett PD, Eckart RE, Bauch TD, et al. Fluoroscopic localization
Imaging of the upper extremity vessels beyond the level of of the femoral head as a landmark for common femoral artery
SCA artery usually requires that a diagnostic catheter be cannulation. Catheter Cardiovasc Interv 2005; 65:205–207.
advanced from the origins of the innominate and left SCA to 7. Rose SE, Kadir S. Arterial anatomy of the upper extremities. In:
the level of the axillary artery. In patients with friendly aortic Kadir S, ed. Normal and Variant Angiographic Anatomy. 1st ed.
arches (i.e., type I), this is usually straightforward and is Philadelphia: W.B. Saunders, 1991:55–96.
achieved by advancing the diagnostic catheter over a long 8. Kurtkoti J, Snow T, Hiremagalur B. Gadolinium and nephrogenic
(300 cm) soft tipped 0.035 in. wire [e.g., Wholey (Mallinckrodt, systemic fibrosis: association or causation. Nephrology (Carlton)
St. Louis, Missouri, U.S.), Magic Torque (Boston Scientific Corp., 2008; 13:235–241.
9. Nagai Y, Hasegawa M, Shinmi K, et al. Nephrogenic systemic
Natick, Massachusetts, U.S.)] that is placed in the brachial artery
fibrosis with multiple calcification and osseous metaplasia. Acta
using the road-mapping function for guidance. The presence of Derm Venereol 2008; 88:597–600.
significant tortuosity in the subclavian artery may occasionally 10. Brosh D, Assali A, Vaknin-Assa H, et al. The ACIST power
require the use of a glidewire (floppy or stiff body) to achieve injection system reduces the amount of contrast media delivered
this maneuver. With complex arch anatomy that necessitates the to the patient, as well as fluoroscopy time, during diagnostic and
use of a Vitek or Simmons catheter for engagement of the left interventional cardiac procedures. Int J Cardiovasc Intervent 2005;
SCA or innominate artery ostia, these catheters need generally to 7:183–187.
[2/7/010/8:59:48] [314–340]
26
Carotid and cerebral angiography

Andrei Pop and Robert D. Safian
INTRODUCTION type depends on the relationship between the point of origin of

Traditionally, cerebral angiography has been performed by the great vessels and the apex of the arch. Although several
neuroradiologists, but the application of percutaneous interven- classifications are available, we rely on a modification of the
tion for chronic brachiocephalic occlusive diseases and acute Myla classification (3) (Fig. 26.1): type 1, in which all three great
stroke has resulted in the increasing involvement of interven- vessels originate from the apex of the arch; type 2, in which the
tional cardiologists. Accordingly, cardiologists are expected to first two great vessels [usually the innominate artery (IA) and
have an understanding of diseases that impact the circulation to left common carotid artery (CCA)] originate below the apex of
the brain, including diseases of the aortic arch, carotid artery, the arch; and type 3, in which all three great vessels originate
subclavian and vertebral artery, and intracranial diseases. This below the apex of the arch. Regardless of which classification
chapter discusses the purpose, specific goals, technique, and is used, the goal is to distinguish “simple” (type 1) from
complications of catheter-based cerebral angiography. Subcla- “complex” (type 2 or 3) arches. Simple arches are conducive
vian artery intervention, extracranial carotid and vertebral inter- to selective cannulation of the great vessels from a femoral
vention, and intracranial and stroke intervention are covered in approach, whereas complex arches make selective cannulation
separate chapters. more difficult due to aortic uncoiling, tortuosity, and the angle
of origin of the great vessels. Aging itself, as well as the long-
term consequences of atherosclerosis and hypertension, lead to
PURPOSE OF ANGIOGRAPHY elongation of the aortic arch, superior displacement of the
Despite the availability of noninvasive techniques such as aortic knob, inferior and posterior displacement of the great
duplex ultrasound, computerized tomography angiography, vessels, and elongation and sharp angulation of the left CCA
and magnetic resonance angiography, invasive angiography (Fig. 26.1). Together, these morphological changes substantially
remains the gold standard for the diagnosis of extra- and increase the technical difficulties of selective cannulation and
intracranial arterial diseases because of superior spatial and angiography, and may partially explain the increased risk of
temporal resolution. In general, invasive angiography may be carotid interventions in octogenarians.
recommended to clarify ambiguous results of noninvasive In contrast to the arch type, the arch configuration refers
imaging, to obtain baseline angiographic evaluation prior to to the usual or anomalous origin of the great vessels. In the
extra- or intracranial interventions, and to assess some patients usual configuration, the IA, left CCA, and the left subclavian
with unusual diseases such as vasculitis, fibromuscular dys- artery (SCA) originate as the first, second, and third great
plasia, dissection, and prior arterial bypass surgery. vessels from the arch, respectively (65% of individuals)
While noninvasive imaging techniques are often used for (Fig. 26.1). Anomalous configurations include common origin
the diagnosis of carotid and vertebral artery diseases, the of the IA and left CCA (so-called bovine configuration, 25%)
quality of these studies is dependent on the acquisition tech- (Figs. 26.2 and 26.3), origin of the left CCA from the proximal
nique and the skill of the physician interpreting the study. IA (7%), separate origin of the left vertebral artery (VA) from
Some noninvasive studies are limited by vessel tortuosity, the arch (0.5%), separate origin of the right SCA from the distal
heavy calcification, or other difficult anatomy. Overreliance arch (arteria lusoria, 0.6%), and origin of the left CCA and left
on noninvasive imaging can result in misdiagnosis or the SCA from a left IA (1.0%). A true bovine configuration is
performance of unnecessary surgery (1,2). It is not unusual extremely rare and consists of a single IA that trifurcates into
for interventions scheduled purely on the basis of noninvasive a right SCA, a single CCA, and a left SCA; the single CCA
imaging to be abandoned or modified once invasive angio- divides into right and left CCA.
graphic images have been obtained. The extent of arch disease should be described in detail,
including the presence of stenosis, degree of calcification and
tortuosity, as well as the presence of aneurysm or ulceration. In
SPECIFIC GOALS OF ANGIOGRAPHY
general, it is important to perform arch aortography prior to
The specific goals of cerebral angiography are to provide a
selective angiography, since the arch type, arch configuration,
detailed assessment of the aortic arch, extracranial brachioce-
and degree of atherosclerosis will influence the need for cath-
phalic vessels, intracranial circulation, and collaterals (Table 26.1).
eter manipulation and catheter selection.
While arch aortography details the anatomy of the
Arch Aortography proximal intrathoracic brachiocephalic circulation (Fig. 26.1),
Every arch study must include an assessment of three key selective angiography is preferred for more detailed assessment
characteristics: the type of aortic arch, the configuration of the of the intrathoracic, cervical, and intracranial segments
great vessels, and the extent of atherosclerosis. The aortic arch (Table 26.2). The purpose of selective angiography is to assess
[2/7/010/8:59:48] [314–340]
CAROTID AND CEREBRAL ANGIOGRAPHY 315
Table 26.1 Specific Goals of Cerebral Angiography

Aortic arch study
l Type of aortic arch (type 1, 2, 3)
l Configuration of the great vessels (usual, anomalous)
l Extent of atherosclerosis
Extracranial circulation
l Intrathoracic segments
l Cervical segments
l Collaterals
Intracranial circulation
l Intracranial vertebral and carotid arteries
l Circle of Willis
l Intracranial branches
l Collaterals
the presence, location, severity, and length of stenosis; normal

reference vessel dimensions; morphological features of the steno-
sis (calcification, thrombus, ulceration, angulation); vessel tor-
tuosity proximal and distal to the stenosis; patency of the major
branches; and the integrity of the intracranial circulation and
collateral pathways. Clinically, selective SCA angiography is
often performed to assess SCA or internal mammary artery
(IMA) graft patency, symptoms of vertebrobasilar insufficiency,
or arm claudication. Subclavian arteriography should be per-
formed in all patients who require cardiac catheterization after
IMA bypass surgery (Fig. 26.4). Subclavian arteriography during
coronary angiography prior to coronary artery bypass surgery
(CABG) is reasonable in patients with known occlusive disease of
the brachiocephalic circulation, cervical or supraclavicular bruits,
discrepant blood pressure between both arms, symptoms sug-
gestive of vertebrobasilar insufficiency, or arm claudication (4).
In most situations, selective SCA angiography provides
sufficient images of the SCA and VA without selective VA
angiography, particularly since most VA stenoses are located at
the VA origin. In selected cases, particularly if VA intervention
is performed, selective VA angiography is required. The SCA is
considered in four sections (Fig. 26.5): proximal (origin of SCA
to origin of VA), mid (segment of SCA involving the origin of
the VA, IMA, and thyrocervical trunk), distal (segment of SCA
distal to thyrocervical trunk and extending up to the axillary
artery), and the axillary artery (Table 26.3). Although angiog-
raphy of the brachial artery circulation is not routine during
arch and cerebral angiography, it is important to have an
understanding of the arterial circulation to the arm and hand.
Selective carotid angiography is performed most often to
assess symptomatic or asymptomatic carotid artery atheroscle-
rosis (Fig. 26.6). Less commonly, carotid angiography is per-
formed to evaluate patency of a carotid-subclavian artery
bypass, fibromuscular dysplasia, or spontaneous carotid dis-
section (Fig. 26.7). The carotid artery is considered in four
segments (Table 26.3): the CCA (intrathoracic and cervical
regions), the external carotid artery (ECA), the cervical internal
carotid artery (ICA), and the intracranial circulation [intra-
cranial ICA, middle cerebral artery (MCA), anterior cerebral
artery (ACA), and posterior communicating artery (PCOM)].
Figure 26.1 Classification of the types of aortic arch. Type 1, in
which all three great vessels originate from the apex of the arch
SPECIFIC VASCULAR TERRITORIES (top). Type 2, in which the innominate artery and left common
External Carotid Artery carotid artery originate below the apex of the arch. Note the extent
of atherosclerosis, ulceration, and stenosis in the origin of all three
After its intrathoracic origin, the CCA bifurcates into the ECA
great vessels (middle). Type 3, in which all three great vessels
and ICA at the level of the C3–C5 interspaces. The ECA is originate below the apex of the arch (bottom).
important because it supplies blood flow to most extracranial
[2/7/010/8:59:48] [314–340]
Table 26.2 Goals of Selective Cerebral Angiography

Vessel assessment
l Proximal and distal tortuosity
l Patency of major branches
Stenosis assessment
l Stenosis severity
l Cause of stenosis (atherosclerosis, fibromuscular dysplasia,
dissection)
l Stenosis morphology
l Normal reference vessel dimensions
Integrity of intracranial circulation
l Circle of Willis
l Collateral pathways
structures in the head and neck (Fig. 26.8), provides numerous

anastomoses to the ICA and VA (Table 26.4), and may supply
blood to intracranial neoplasms and vascular malformations.
Some of these diseases may be outside the expertise of interven-
tional cardiologists, but may be encountered during angiogra-
phy for other reasons. For physicians interested in carotid
Figure 26.2 Bovine configuration of the aortic arch. artery stenting, the anatomy of the ECA is important for use
of proximal embolic protection devices. Branches of the ECA
include five anterior and three posterior arteries, which are
visualized best in the lateral projection. Variations in the origin,
size, and distribution of ECA branches are fairly common
(Fig. 26.9), but anomalies of the ECA are rare.
Internal Carotid Artery

In most patients, the ICA has no extracranial branches and is
longer than the ECA. Aberrant origin of the occipital artery or
ascending pharyngeal artery from the ICA is very rare. The ICA
provides most of the blood flow to the cerebral hemispheres,
and vascular diseases of the ICA and its branches are important
causes of serious morbidity, disability, and mortality. There
are several classifications of the segments of ICA (Table 26.5),
but we prefer a simplified classification based on cervical
(extracranial), petrous (intracranial but extradural), cavernous
(within the cavernous sinus), and cerebral (intradural) seg-
ments (Figs. 26.10 and 26.11). The cervical segment of the ICA
is the most common site for carotid atherosclerosis, and con-
sists of the carotid bulb and the ascending ICA (Fig. 26.12). The
proximal cervical ICA is anatomically related to the cervical
sympathetic ganglion, the vagus nerve, and the hypoglossal
nerve, which may explain vasovagal and vasodepressor
responses during ICA stimulation. The petrous segment of
the ICA has a characteristic appearance that includes a vertical
segment and sharp horizontal segment (upside-down “L” or
genu), whereas the cavernous segment has two sharp genus
(short vertical segment, longer horizontal segment, and a short
vertical segment) (Figs. 26.10 and 26.11). Operators who per-
form carotid artery stenting will commonly position a distal
embolic protection device in the upper cervical segment, and
the guidewire tip will often reside in the petrous segment,
avoiding the cavernous and cerebral segments (Fig. 26.12).
Figure 26.3 Anomalous configuration of the aortic arch in which
the innominate artery and left common carotid artery share a com- Circle of Willis
mon infundibulum (bovine configuration). Note the severe stenosis It is important to have an understanding of the normal anat-
of the left subclavian artery and retrograde filling of the left vertebral omy, common anatomic variations, and vascular anomalies of
artery (arrow) from the right vertebral artery (subclavian steal). the circle of Willis. The circle of Willis provides blood flow to
the anterior, posterior, left, and right portions of the brain, and
[2/7/010/8:59:48] [314–340]
Figure 26.4 Subclavian and coronary steal after coronary artery bypass surgery. (A) Selective left subclavian arteriogram (308 RAO
projection) demonstrates total occlusion of the left subclavian artery, and no antegrade filling of the left vertebral artery or the left internal
mammary artery (LIMA) graft to the LAD (left). After initial angioplasty, there is high-grade residual stenosis, extending up to the origin of the
left vertebral artery (right). Note the lucency at the origin of the left vertebral artery (arrowhead), which represents bidirectional blood flow and
should not be mistaken for thrombus. (B) After successful stenting, there is prompt antegrade flow in the left subclavian and vertebral arteries
(left, arrowhead), as well as the internal mammary bypass graft (arrow). Selective angiography demonstrates patency of the LIMA graft to the
LAD (right). Abbreviation: RAO, right anterior oblique.
Figure 26.5 Selective angiography of the left subclavian artery. RAO projection (left), LAO projection (right). The proximal (1), mid (2), and
distal segments (3) are evident; the axillary artery is not shown. Abbreviations: RAO, right anterior oblique; LAO, left anterior oblique.
[2/7/010/8:59:48] [314–340]
Table 26.3 Sections of the Carotid and Subclavian Arteries

Section of the subclavian artery
l Proximal (origin of SCA to origin of VA)
l Mid (SCA involving VA, IMA, and TCT)
l Distal (Distal to TCT to AA)
l Axillary (AA)
Sections of the carotid artery
l CCA (intrathoracic and cervical)
l ECA
l Cervical ICA
l Intracranial ICA (ICA, MCA, ACA, PCOM)
Abbreviations: SCA, subclavian artery; VA, vertebral artery; IMA,
internal mammary artery; TCT, thyrocervical trunk; AA, axillary
artery; CCA, common carotid artery; ECA, external carotid artery;
ICA, internal carotid artery; MCA, middle cerebral artery; ACA,
anterior cerebral artery; PCOM, posterior communicating artery.
is a key source of collateral blood flow in occlusive diseases of

the intra- and extracranial circulation. A complete circle of
Willis is present in 50% of individuals, and consists of a nine-
sided polygon (nonagon) and the basilar artery (Fig. 26.13). The
complete “circle” includes the anterior communicating artery
(ACOM), two precommunicating anterior cerebral arteries (A1), Figure 26.6 Selective bilateral carotid artery angiography in a
two intracranial ICA; two precommunicating posterior cerebral 68-year-old woman with right hemispheric TIA. Results of a carotid
arteries (P1); and two PCOM (Table 26.6). The postcommuni- duplex ultrasound study were ambiguous due to ischemic cardiomy-
cating anterior (A2) and posterior (P2) cerebral arteries and the opathy, low cardiac output, and aortic valve disease. Selective left
MCA are not part of the circle of Willis. The entire circle of carotid angiography (left) demonstrated an unexpected severe steno-
Willis is rarely identified during a single-vessel angiogram, so sis in the distal common carotid artery (CCA) with prominent ulcer-
ation and eccentricity. Note the catheter position in the mid-CCA
visualization of the individual components depends on the
(arrowhead), and moderate disease and prominent tortuosity of the
extent of the cerebral angiogram. cervical and petrous segments of the left internal carotid artery (ICA)
Anatomic variations in the circle of Willis are very com- (arrows). Selective right carotid angiography (right) demonstrated
mon (*50%), particularly aplasia or hypoplasia of PCOM and severe ulcerated stenosis of the ICA with calcification, later treated
P1 (Fig. 26.14), and may impair potential collateral pathways. by successful carotid artery stenting with embolic protection. Note that
Individuals with ipsilateral aplasia or severe hypoplasia of A1 the diagnostic catheter is in the proximal right CCA, resulting in reflux
and P1 have an isolated cerebral hemisphere and cannot receive of contrast into the right subclavian and vertebral arteries (arrow).
collateral flow from the vertebrobasilar system or from the
Figure 26.7 Selective left carotid artery angiogram (left) in a 72-year-old woman with severe stenosis in the proximal carotid-to-subclavian
bypass (arrow), which explained a loud supraclavicular bruit, thrill, and discrepant blood pressures in both arms. Selective left carotid artery
angiogram (middle) in a young woman with fibromuscular dysplasia (FMD). Although the angiogram is highly reliable for identifying FMD, it is
less reliable for assessing stenosis severity. Note the beaded aneurysms characteristic of medial fibroplasia. She did not have intracranial
aneurysms. Selective left carotid artery angiogram (right) in a young woman who presented with severe neck pain and partial Horner’s
syndrome. Note the long spiral spontaneous dissection of the cervical and petrous segments of the internal carotid artery.
[2/7/010/8:59:48] [314–340]
Table 26.4 Major Anastomoses of the External Carotid Artery

ECA branch Destination
Superior thyroid artery SCAa
Ascending pharyngeal arteryb ICA, ECA, SCA, VA
Lingual artery ECA
Facial artery ICAc, ECA
Occipital arteryb ECA, SCA, VA
Posterior auricular arteryb ECA
Superior temporal artery ICAc, ECA
Inferior maxillary artery ICA, ECA
a
Collateral to ipsilateral SCA via inferior thyroid artery and thyrocer-
vical trunk, or to contralateral SCA via opposite superior thyroid
artery.
b
These branches usually have posterior origin from the ECA.
c
Via ophthalmic artery.
Abbreviations: SCA, subclavian artery, ICA, internal carotid artery,
ECA, external carotid artery, VA, vertebral artery.
artery represents the distal continuation of the main ACA

(Fig. 26.19). Aplasia or hypoplasia of A1 is observed in 12% of
individuals, but other anatomic variations are rare. Two ACA
anomalies (infraoptic origin, azygous ACA) are associated with
intracranial aneurysms and AVM, but bihemispheric ACA is
not (Table 26.7).
Middle Cerebral Artery

The MCA and its branches supply most of the blood flow to the
superior, lateral, and central portions of the brain (Figs. 26.15
and 26.17–26.21). The most common classification of the MCA
includes four segments (M1, M2, M3, M4); M2 and M3 are
conduit segments between M1 and M4, and usually do not
Figure 26.8 Schematic illustration of branches of the external provide significant branches (Table 26.10). Perforating branches
carotid artery. 1, Superior thyroid artery; 2, lingual artery; 3, facial arise from M1 (lateral lenticulostriate branches to basal ganglia
artery; 4, maxillary artery; 5, superior temporal artery; 6, occipital
and internal capsule), and cortical branches arise from M1
artery; 7, posterior auricular artery; 8, ascending pharyngeal artery.
Note that the internal carotid artery has been cut away for clarity.
(anterior temporal artery to anterior and inferior temporal
lobe) and from M4 (supplying large sections of the temporal,
frontal, parietal, and occipital lobes) (Tables 26.11 and 26.12;
Figs. 26.15 and 26.21). A common anatomic variation is early
MCA bifurcation (20% of individuals), but other variations are
rare. MCA anomalies such as fenestrations, duplications, and
contralateral ICA; these patients are especially prone to cerebral
accessory MCA are associated with intracranial aneurysm and
ischemia if they develop proximal occlusive disease. Anomalies
AVM (Table 26.7).
of the circle of Willis and other intracranial arteries are uncom-
mon. However, duplication and fenestration of intracranial
vessels are associated with intracranial aneurysms and arterio- Vertebrobasilar Circulation
venous malformations (AVM) (Table 26.7). The vertebrobasilar system provides virtually all of the blood
flow to the posterior circulation, including the pons, medulla,
Anterior Cerebral Artery midbrain, and cerebellum (Figs. 26.15–26.17). The vertebroba-
The ACA and its branches supply the majority of blood flow silar system consists of extracranial (vertebral arteries) and
to the anterior, medial, and anterobasal portions of the brain intracranial [vertebral arteries, basilar artery, posterior cerebral
(Figs. 26.15–26.20). The most common classification of the ACA arteries (PCAs)] components (Table 26.13). The most common
includes three segments (A1, A2, A3), which provide perforat- classification of the VA includes four segments (V1, V2, V3, V4)
ing and cortical branches to the brain (Table 26.8). Perforating depending on the relationship to the cervical transverse fora-
branches usually arise from A1 and A2 (basal ganglia, internal mina and foramen magnum (Table 26.14; Figs. 26.22 and 26.23).
capsule, corpus callosum) and cortical branches usually arise Cervical branches originate from the V1 segment (branches to
from A2 (ventral medial frontal lobe) and A3 (corpus callosum, the spinal cord and skeletal muscles of the head and neck);
medial cerebral hemisphere, cortical convexity) (Table 26.9). meningeal branches originate from V2 and V3; and intracranial
Most of the penetrating and cortical branches can be readily branches originate from V4 (branches to the cerebellum, pons,
identified by selective ipsilateral carotid angiography in ante- medulla, and spinal cord). Anatomic variations in the size of
roposterior (AP)-cranial and lateral projections. The pericallosal the VA are common; although the left VA is large and
[2/7/010/8:59:48] [314–340]
Figure 26.9 Selective angiography of the left carotid artery, demonstrating the distal common carotid artery (CCA), internal carotid artery
(ICA), and branches of the external carotid artery (ECA). (A) Note the severe stenoses at the origin of the ICA and ECA. (B) Note the variation
in the ECA in which the lingual and facial arteries originate from a common trunk (TR), rather than as separate branches from the ECA.
Abbreviations: STA, superior thyroid artery; L, lingual artery; F, facial artery; MAX, maxillary artery; TA, temporal artery; OCC, occipital artery;
APA, ascending pharyngeal artery; PAA, posterior auricular artery.
Table 26.5 Cervical and Intracranial Segments of the Internal Carotid Artery
Region Segment Description Notable branches
Cervical C1 Cervical Carotid bulb and ascending cervical ICA None
(extracranial)
Petrous C2 Petrous ICA contained within petrous bone; characteristic genu Vidian artery, caroticotympanic artery
C3 Lacerum ICA travels above FL between petrous bone and PL None
ligament (extradural, intracranial)
Cavernous C4 Cavernousa Only artery that courses through a vein (cavernous sinus); Posterior trunk,b lateral trunk, capsular
characteristic genu (extradural, intracranial) arteries (medial)
C5 Clinoida Short wedge of ICA in terminal cavernous sinus (interdural) None
Cerebral C6 Ophthalmica ICA exits cavernous sinus (intradural) Ophthalmic artery, superior
hypophyseal artery
C7 Communicating Origin of PCOM to ACA/MCA bifurcation PCOM, anterior choroidal artery
a
C4, C5, C6, carotid siphon.
b
May cause a pituitary blush (meningohypophyseal artery).
Abbreviations: ICA, internal carotid artery; FL, foramen lacerum; PL, petrolingual ligament; PCOM, posterior communicating artery; ACA,
anterior cerebral artery; MCA, middle cerebral artery.
[2/7/010/8:59:48] [314–340]
Figure 26.10 Schematic representation of the classification of the

segments of the internal carotid artery in the lateral and AP
projections. The simplified classification designates four segments,
including cervical (C1), petrous (C2, C3), cavernous (C4, C5), and
cerebral (C6, C7) segments.
dominant in most individuals, the right VA is dominant in 25%

of cases (Table 26.7). The most common VA anomaly is direct
origin from the arch rather than from the SCA, which occurs in
5% of individuals. Although anomalous origin of the VA is not
associated with serious intracranial diseases, duplications and
fenestrations of the VA are associated with fused vertebrae,
intracranial aneurysms, and AVM (Table 26.7).
The basilar artery is a large midline artery that is formed
by the merging of the left and right VA, and extends to the
bifurcation of both PCAs (Figs. 26.13, 26.16, 26.17, 26.22, and
26.23). Although only 3 to 4 mm in diameter and 30 mm in
length, the basilar artery has several important branches,
including pontine perforators and two of three major cerebellar
arteries [anterior inferior cerebellar artery (AICA), superior
cerebellar artery] (Table 26.13; Figs. 26.16, 26.17, and 26.23).
The third major artery to the cerebellum [posterior inferior
cerebellar artery (PICA)] is a branch of the VA. Anatomic
variations of the basilar artery are not common, but a common
AICA-PICA trunk occurs in 10% of individuals. Basilar artery
fenestrations and duplications are associated with intracranial
Figure 26.11 Selective angiography of the left internal carotid
aneurysms (Table 26.7).
artery (ICA) in the AP (top) and lateral projection. Note that the
segments of the ICA are easier to discern in the lateral projection
Posterior Cerebral Artery due to less foreshortening of the cavernous segment.
The PCA and its branches supply most of the blood flow to the
posterior and posterobasal portions of the brain and brainstem
(Table 26.15; Figs. 26.15–26.17 and 26.24). The most common
classification of the PCA includes four segments (P1, P2, P3, P4), and midbrain; choroidal branches arise from P2 and provide
which provide perforating branches, branches to the choroid blood flow to the choroid plexus and basal ganglia; and cortical
plexus and ventricles, and cortical branches to the brain branches arise from P2 (providing blood flow to the temporal
(Table 26.16). Perforating branches usually arise from P1 and lobe) and P4 (providing blood flow to the temporal, parietal,
P2, providing blood flow to the internal capsule, basal ganglia, and occipital lobes) (Table 26.16; Fig. 26.24). Normal variations
[2/7/010/8:59:48] [314–340]
ANGIOGRAPHIC TECHNIQUE
Arterial Access
Cerebral angiography and intervention are generally per-
formed from a femoral artery approach, but radial (5), brachial
(6), and ulnar (7) approaches have been used when femoral
access is difficult. Local anesthesia is achieved with 1% lido-
caine, and light conscious sedation may be employed with
midazolam (1 mg intravenously). We recommend a 5-Fr sheath
and 5-Fr catheters for diagnostic angiography, and intra-arterial
heparin (1500–2000 IU) in all patients (Table 26.18).
Pressure Monitoring and Imaging

As is true for left heart catheterization and coronary angiog-
raphy, we recommend intra-arterial pressure monitoring from
the catheter tip, using a standard three- or four-port manifold
(Table 26.18). Meticulous attention must be paid to monitoring
catheter tip pressure during cannulation of the great vessels. If
a dampened pressure waveform is observed the catheter
should be repositioned. Catheters that cannot be aspirated
should never be flushed in the arch or great vessels, but should
be removed and reflushed outside the patient. The highest
quality images are obtained with digital subtraction angiogra-
phy (DSA), a power injector, and isosmolar contrast such as
iodixanol. Patient cooperation is essential, so instructions
should be provided on breath-hold and avoidance of move-
ment during angiography. Hand injections are useful to verify
catheter position and alignment, but generally result in sub-
optimal vessel opacification, contrast streaming, poor images of
the intracranial circulation, and higher radiation doses to the
angiographer (Fig. 26.26) (8).
Arch Study
A formal arch aortogram [408 left anterior oblique (LAO)]
should be performed in all patients unless the arch type and
configuration have been previously defined by noninvasive or
invasive imaging (Table 26.18; Figs. 26.1–26.3). For patients who
Figure 26.12 Selective angiography of the left carotid artery during require only left SCA arteriography to evaluate patency of an
carotid artery stenting. Note the severe stenosis in the proximal IMA graft, a formal arch study may not be necessary. Several
internal carotid artery (ICA), just distal to the carotid bulb (CB). A catheters may be used for arch aortography, but the ones
distal embolic protection device is positioned in the upper cervical used most often are the pigtail and tennis racquet catheters
segment (large arrow), and the guidewire tip is in the proximal (Fig. 26.27). The side-hole arrangement on the tennis racquet
petrous segment of the ICA (short arrow). catheter minimizes contrast in the ascending aorta and pro-
motes contrast flow from the catheter shaft directly into the
great vessels. High-quality DSA images will be obtained by
instructing the patient to turn the head to the left, and by
injecting 25 to 30 cc contrast over two seconds (600 psi), with a
field of view of 34 to 42 cm (Table 26.18).
of the PCA are quite common, including P1 hypoplasia in 20%
of individuals, which is associated with persistent fetal origin of
the PCA (Table 26.7; Fig. 26.25). Anomalies of the PCA are quite Selective Cannulation
rare. Several catheters and techniques are available for selective
cannulation of the great vessels and it is useful to become
familiar with more than one (Tables 26.19 and 26.20). The arch
Intracranial Aneurysms and Other Conditions study serves as a roadmap before selective cannulation and
Cardiologists are rarely involved in the angiographic evaluation minimizes the need for catheter manipulation in the arch. If an
and treatment of patients with intracranial aneurysms or tumors. arch study is not available, the tracheal air stripe (408 LAO
However, aneurysms may be identified incidentally during projection) is a useful landmark for the origin of the great
cerebral angiography for other reasons (Fig. 26.20), so cardiolo- vessels: in many patients the IA is just proximal (left), the left
gists should have an understanding of the incidence, distribu- CCA is in the middle, and the left SCA is just distal (right) to air
tion, and outcomes of intracranial aneurysms (Table 26.17). stripe (Fig. 26.28). If rapid cannulation cannot be achieved, it is
Other incidental findings might include neovascular tumor better to perform an arch study than to persist with prolonged
blush, an arteriovenous fistula, or an AVM. catheter manipulation. The most common reasons for failed
[2/7/010/8:59:48] [314–340]
Figure 26.13 Schematic illustration of the circle of Willis. (A) A complete circle of Willis is actually a nine-sided polygon plus the basilar
artery. (B) A complete circle of Willis with associated major branches. See text for details. Abbreviations: ACOM, anterior communicating
artery; A1, precommunicating anterior cerebral artery(ACA); A2, postcommunicating ACA; ICA, internal carotid artery; MCA, middle cerebral
artery; PCOM, posterior communicating artery; P1, precommunicating posterior cerebral artery (PCA); P2, postcommunicating PCA; BA,
basilar artery; VA, vertebral artery; SCeA, superior cerebellar artery; AICA, anterior inferior cerebellar artery; PICA, posterior inferior
cerebellar artery.
Table 26.6 Components of the Complete Circle of Willisa classified as simple or complex, depending on the primary
and secondary curves (Fig. 26.29). Simple catheters include
Component Number Judkins right (JR), no torque right (NTR), Headhunter, jugular
ACOM 1a bulb (JB), and vertebral catheters. In general, simple catheters
A1 2a have simple primary and secondary curves, are easier to use,
ICA 2a require less contact with the outer curvature of the arch, and
PCOM 2a are well suited for relatively simple arch configurations. Selec-
P1 2a tive engagement with a simple catheter involves retraction of
BA 1 the catheter and counterclockwise rotation (Fig. 26.30). If the tip
a
Nine-sided polygon plus the BA. is too long, a shorter curve may be used. In contrast, complex
Abbreviations: ACOM, anterior communicating artery; A1, precom- catheters, such as Simmons and Vitek (Cook, Inc., Bloomington,
municating anterior cerebral artery; ICA, distal intracranial internal Indiana, U.S.), have complex curves, require more manipula-
carotid artery; PCOM, posterior communicating artery; P1, precom- tion and skill, result in more contact with the outer curvature of
municating posterior cerebral artery; BA, basilar artery.
the arch, and are suited for many simple and complex arch
configurations. Among complex catheters, the Vitek is the
easiest to use: selective engagement involves positioning the
selective cannulation of a great vessel are complex arch config- catheter distal to the left SCA then rotating and advancing it
urations and anomalous origins; both are readily identified by until the tip points superiorly. Each great vessel may be
an arch study. selectively engaged by advancing the Vitek catheter from the
Catheter selection is largely based on operator prefer- left SCA to the left CCA to the IA (Fig. 26.30). The Simmons
ence, arch type, and the presence of anomalous origins catheter requires more manipulation, and must be formed first
(Table 26.19). Catheters for selective angiography are often in either the left SCA or the ascending aorta (Fig. 26.30). Most
[2/7/010/8:59:48] [314–340]
Figure 26.14 Schematic illustration of common variations in the circle of Willis. (A) Aplasia or severe hypoplasia of P1 [precommunicating
posterior cerebral artery (PCA)]; (B) aplasia or severe hypoplasia of A1 (precommunicating anterior cerebral artery [ACA]); (C) aplasia or
severe hypoplasia of P1 and persistent fetal PCA; (D) isolated cerebral hemisphere, due to aplasia or severe hypoplasia of A1 and P1 (with
persistent fetal PCA); (E) aplasia or severe hypoplasia of the anterior communicating artery (ACOM).
cardiologists will prefer the Vitek catheter, which handles in a simple curve catheter, and distal to proximal with a Vitek
fashion that is very similar to a left Amplatz catheter; gentle catheter (Fig. 26.30).
retraction results in deeper seating, while advancement results Once the great vessel has been selectively engaged and if
in disengagement. In our experience, all great vessels have been the catheter provides good coaxial alignment without pressure
successfully engaged with a simple curve catheter and/or a damping, selective angiography may be performed with the
Vitek catheter. For a diagnostic angiogram, the sequence of same catheter, preferably using a power injector. If there is
selective cannulation is generally proximal to distal with a pressure damping, suboptimal position, or noncoaxial alignment
[2/7/010/8:59:48] [314–340]
Table 26.7 Normal Variations and Anomalies of the Cerebral Artery Circulation
Vessel Variation Incidence (%) Anomaly Incidence (%)
VA Right dominant 25 Aortic origin 5
AICA-PICA trunk 10 Fenestrationa <1
Duplicationa <1
PICA from V1 10
BA – Fenestrationa <1
Duplicationa <1
PCA P1 hypoplasia/aplasiab 20 –
MCA Early bifurcation 20 Accessory MCAa 2
Fenestrationa <1
Duplicationa 2
ACA A1 hypoplasia/aplasia 12 Infraoptic origina <1
Bihemispheric ACA 4
Azygous ACAa 2
PCOM Hypoplasia/aplasia 5 –
ACOM Hypoplasia/aplasia 5 –
a
Associated with intracranial aneurysm and AVM.
b
Associated with persistent fetal origin of PCA.
Abbreviations: VA, vertebral artery; BA, basilar artery; PCA, posterior cerebral artery; MCA, middle cerebral artery; ACA, anterior cerebral
artery; PCOM, posterior communicating artery; ACOM, anterior communicating artery; AICA, anterior inferior cerebellar artery; PICA,
posterior inferior cerebellar artery; AVM, arteriovenous malformation.
Figure 26.15 Schematic illustration of the vascular distributions of the anterior cerebral artery (ACA), middle cerebral artery (MCA), and
posterior cerebral artery (PCA), shown in the lateral, medial, superior, and basal views.
[2/7/010/8:59:48] [314–340]
Figure 26.16 Schematic illustration of the intracranial circulation,

medial view. Note that in this view, about one-third of blood flow is
derived from the vertebrobasilar system, posterior cerebral artery
(PCA), and their branches (1–8, 16–20), and two-thirds from the
anterior cerebral artery (ACA) and its branches (9–15). 1, Perforator
branches; 2, choroidal and ventricular branches; 3, anterior tempo-
ral artery (cut away); 4, posterior temporal artery; 5, medical
occipital artery; 6, parieto-occipital artery; 7, calcarine artery; 8, Figure 26.17 Schematic illustration of the intracranial circulation,
splenial artery; 9, postcommunicating ACA; 10, orbitofrontal artery; basal view. Note that in this view, about 50% of blood flow is derived
11, frontopolar artery; 12, callosomarginal artery; 13, pericallosal from the vertebrobasilar system, posterior cerebral artery and their
artery; 14, splenial artery; 15, parietal artery; 16, vertebral artery; 17, branches (1–13); 35% from the anterior cerebral artery and its
basilar artery and perforators; 18, superior cerebellar artery; 19, branches (14–16); and 15% from the middle cerebral artery and
anterior inferior cerebellar artery; 20, posterior inferior cerebellar its branches (17). 1, Precommunicating PCA; 2, postcommunicating
artery. Abbreviation: ICA, internal carotid artery. PCA; 3, P4 (calcarine) segment of PCA; 4, anterior temporal artery;
5, posterior temporal artery; 6, parieto-occipital artery; 7, calcarine
artery; 8, basilar artery and perforators; 9, superior cerebellar artery;
10, anterior inferior cerebellar artery; 11, posterior inferior cerebellar
artery; 12, anterior spinal artery; 13, vertebral artery; 14, anterior
communicating artery; 15, precommunicating ACA; 16, orbitofrontal
that cannot be corrected by gentle catheter manipulation, the artery; 17, middle cerebral artery.
catheter may be exchanged over a guidewire for a multipurpose
angiographic catheter [or modified tennis racquet (Meditech/
Boston Scientific, Watertown, Massachusetts, U.S.)], if necessary
(Tables 26.19 and 26.20; Figs. 26.27 and 26.29).
Selective Angiography of the IA Selective Angiography of the Right SCA

The LAO arch study (Figs. 26.1–26.3) provides nice images of Depending on the clinical circumstances, selective right SCA
the ostia and proximal IA, left CCA, and left SCA, but is less angiography may be performed after advancing the IA catheter
useful for visualizing the IA bifurcation into the right CCA and into the right SCA (Table 26.20). This may be accomplished by
right SCA (Fig. 26.31). Imaging the IA in the LAO and AP clockwise rotation and advancement of a simple catheter using
projections may result in overlap of the right VA and right fluoroscopy, pressure monitoring, and test injections by hand.
CCA, making selective cannulation of the SCA and CCA more If necessary, a J-tip guidewire may be positioned in the axillary
difficult. In contrast, selective angiography of the IA in a 308 artery, followed by advancement of any suitable coaxial cath-
right anterior oblique (RAO) projection will clearly demon- eter into the right SCA. To study the right IMA, an IMA or
strate the IA bifurcation and separate the right VA from the Bartorelli catheter may be used for selective cannulation and
CCA (Table 26.20; Fig. 26.31). This angiogram may be per- angiography. If the purpose is to image the VA and PCA, it is
formed by hand injection. The purpose of the IA angiogram is best to perform right SCA with a power injector, or to selec-
to provide a better roadmap for selective cannulation and tively image the right VA with a vertebral or JR catheter. The
angiography of the right SCA and right CCA, and should not best angiographic projections depend on the vessel of interest
be used for images of the right VA, right ICA, and intracranial (Table 26.20); cine angiography rather than DSA is preferred for
circulation, due to vessel overlap and poor opacification. the images of IMA and coronary artery using a hand injection.
[2/7/010/8:59:48] [314–340]
Selective Angiography of the Right CCA

From the IA, selective right CCA angiography may be performed
after advancing the IA catheter into the right CCA. This is most
easily accomplished in the RAO projection by clockwise rotation
(from the IA) and advancement of a simple catheter using fluo-
roscopy, pressure monitoring, and test injections (Fig. 26.31). If a
complex catheter was used to engage the IA, gentle retraction of
the catheter and clockwise rotation will point the catheter tip into
the right CCA. If these maneuvers do not work, a J-tip guidewire
can be advanced into the mid-CCA followed by advancement of
any suitable coaxial catheter. For images of the carotid bifurcation
and intracranial circulation, hand injections usually result in
contrast streaming and suboptimal image quality, so a power
injector is preferred (Fig. 26.26). If the catheter position is not ideal
or the catheter recoils into the arch, the catheter can be reposi-
tioned with a guidewire or exchanged for a more coaxial catheter,
such as a modified tennis racquet, multipurpose or glide catheter
(Figs. 26.27 and 26.29). In most patients, the best images of the
right CCA and cervical ICA are obtained in AP, lateral, and
ipsilateral oblique (RAO) projections (Figs. 26.20 and 26.32).
Depending on the individual patient, contralateral oblique
Figure 26.18 Schematic illustration of the intracranial circulation (LAO) and cranial or caudal angulation may be useful to image
(anterior, coronal view), including a portion of the circle of Willis highly eccentric stenoses, straighten bends, or eliminate over-
(1–4). The left side demonstrates the middle cerebral artery (MCA) lying branches of the ECA. For the intracranial circulation, the
and its branches (5–10) and the right side demonstrates the anterior
best images are AP-cranial and lateral projections (Figs. 26.20,
cerebral artery (ACA) and its branches (3, 11–13). ACA is cut away
on the left, MCA is cut away on the right. 1, Internal carotid artery; 2,
26.25, and 26.32). In some patients, highly angulated or tortuous
posterior communicating artery; 3, precommunicating ACA; 4, ante- sections of the cerebral ICA may appear aneurysmal, in which
rior communicating artery; 5, M1 (horizontal) segment of MCA; 6, case an ipsilateral (RAO) caudal projection may prove very
lateral lenticulostriate branches; 7, M2 (insular) segment of MCA; 8, useful.
M3 (opercular) segment of MCA; 9, M4 (cortical) segments of MCA;
10, anterior temporal artery; 11, postcommunicating ACA; 12, Selective Angiography of the Left CCA
medial lenticulostriate branches; 13, recurrent artery of Heubner. In the presence of simple or complex arch anatomy, the left
CCA may be cannulated in a 408 LAO projection with the same
Figure 26.19 Selective angiography of the left carotid artery, demonstrating the intracranial circulation in the lateral (left) and AP (right)
projections. Note that the branches of the anterior cerebral artery (ACA) and middle cerebral artery (MCA) are superimposed in the lateral
projection. Abbreviations: OFA, orbitofrontal artery; FPA, frontopolar artery; CMA, callosomarginal artery; PeCA, pericallosal artery; M1,
horizontal segment of MCA; M2, insular segment of MCA; M3, opercular segment of MCA; PPA, posterior parietal artery (M4 segment of
MCA); AA, angular artery (M4 segment of MCA); A1, precommunicating ACA; A2, postcommunicating ACA; LLSA, lateral lenticulostriate
branches.
[2/7/010/8:59:48] [314–340]
Figure 26.20 Selective angiography of both carotid arteries and intracranial circulation. (A) Selective right carotid angiography demon-
strates the right internal carotid artery (ICA) (left 408 LAO projection) and right intracranial circulation (right, lateral projection). Note the severe
stenosis in the ICA, tortuosity of the petrous and cavernous ICA, absence of opacification of the right anterior cerebral artery (ACA) and its
branches (see Fig. 26.19 for comparison), and a small aneurysm of the cavernous segment (arrow). (B) Composite of selective right (left) and
left (right) intracranial circulation in the AP-cranial projection, confirms absence of opacification of the right ACA (see Fig. 26.19 for
comparison), and extensive left-to-right crossover via the ACA and anterior communicating artery. Abbreviations: M1, horizontal segment of
MCA; M2, insular segment of MCA; AA, angular artery; M3, opercular segment of MCA.
catheter that was used to engage the IA. In some patients, the used to engage the IA but is too long to engage the left CCA,
curve of a simple catheter is too long to engage the left CCA, greater success will be achieved with a JR3.5, NTR, or AR-1
and a catheter with a more horizontal orientation (shorter) may catheter. A Vitek catheter will also successfully engage the left
work better (Tables 26.19 and 26.20). For example, if a JR4 is CCA in virtually all patients with simple or complex arch
[2/7/010/8:59:48] [314–340]
Table 26.8 Classification of the Anterior Cerebral Artery

Segment Region Branches Projection
A1 Horizontal (precommunicating; horizontal segment of ACA between ICA Perforating AP-C
and ACOM)
A2 Vertical (postcommunicating; vertical segment of ACA after ACOM) Perforating and cortical AP-C, LAT
A3 Callosal (distal segment of ACA that extends around corpus callosum) Cortical LAT
AP-C, anteroposterior with cranial angulation; LAT, lateral; ICA, internal carotid artery; ACOM, anterior communicating artery; ACA, anterior
cerebral artery.
Table 26.9 Perforating and Cortical Branches of the Anterior Cerebral Artery
Vessel Perforating branch Territory Cortical branch Territory
A1a MLSA Basal ganglia, internal – –
RAHb capsule, corpus callosum
Callosal
A2 RAHa,b Basal ganglia, internal Orbitofrontalc Ventral, medial frontal lobe and olfactory bulb
capsule Frontopolarc
A3c – – Pericallosal Corpus callosum, 2/3 of medial cerebral
Callosomarginal hemisphere, strip of cortex over superior
Parietal convexity
Splenial
a
Best seen in AP-C projection.
b
RAH originates from A1 in 40% of patients, RAH from A2 in 60% of patients.
c
Best seen in LAT projection.
Abbreviations: MLSA, medial lenticulostriate arteries; RAH, recurrent artery of Heubner; AP-C, anteroposterior with cranial angulation; LAT,
lateral; ACA, anterior cerebral artery.
Table 26.10 Classification of the Middle Cerebral Artery

Segment Region Branches Projection
M1 Horizontal (ICA Perforating AP-C
bifurcation to genu) and cortical
M 2a Insular (superior None AP-C, LAT
course after genu)
M 3a Opercular (top of None AP-C, LAT
insula to lateral end
of Sylvian fissure)
M4 Cortical (lateral Cortical AP-C, LAT
surface of Sylvian
fissure to cortical
surface of cerebral
hemisphere)
a
M2 and M3 are conduit vessels between M1 and M4, and do not
have major penetrating or cortical branches.
Abbreviations: ICA, internal carotid artery; AP-C, anteroposterior
with cranial angulation; LAT, lateral.
Table 26.11 Perforating and Cortical Branches of the Middle

Cerebral Artery
Perforating Cortical
Vessel branch Territory branch Territory
Figure 26.21 Schematic illustration of the middle cerebral artery M1 LLSA a
Basal ATA a,b
Anterior, medial,
(MCA) and branches of the M4 segment in the lateral view. In this ganglia and inferior
view of the lateral surface of the brain, the MCA branches are temporal lobe
considered anterior (A, 1–3), central (C, 4–6), or posterior (P, 7–11). M4 – – See Table See Table 26.12
The central sulcus (C) and Sylvian fissure (S) are important land- 26.12
marks for MCA branches. 1, Anterior temporal artery; 2, orbitofrontal a
Best seen in AP-C projection.
artery; 3, prefrontal artery; 4, precentral sulcus artery; 5, central b
Best seen in LAT projection.
sulcus artery; 6, postcentral sulcus artery; 7, posterior parietal
Abbreviations: LLSA, lateral lenticulostriate arteries; ATA, anterior
artery; 8, angular artery; 9, temporo-occipital artery; 10, posterior
temporal artery; AP-C, anteroposterior with cranial angulation; LAT,
temporal artery; 11, middle temporal artery.
lateral.
[2/7/010/8:59:48] [314–340]
Table 26.12 Cortical Branches of M4

Region Location Branch Territory
Anterior Frontotemporal Anterior temporal Anterior, medial, and inferior temporal lobe
Temporopolar
Orbitofrontal Anterior frontal lobe
Prefrontal
Central Frontoparietal Precentral sulcus Posterior frontal, anterior parietal lobes
Central sulcus Superior parietal lobe
Postcentral sulcus Mid-parietal lobe
Posterior Parieto-occipital-temporal Posterior parietal Posterior parietal lobe
Angulara Posterior parietal and occipital lobes
Temporo-occipital Posterior temporal and occipital lobes
Posterior temporal Middle temporal lobes
a
Major terminal branch of the middle cerebral artery.
Table 26.13 Branches of the Vertebrobasilar System

Vertebral arterya
l Cervical branches (originate from V1 segment)
Muscular branches
Spinal branches
l Meningeal branches (originate from distal V2 and V3 segments)
Anterior meningeal artery (small; originate from distal V2
segment)
Posterior meningeal artery (large; originate from V3 segment)
l Intracranial branches (originate from V4 segment)
Posterior spinal artery
Anterior spinal artery
Perforating branches
Posterior inferior cerebellar artery (PICA)
Basilar arteryb
Pontine perforators
Anterior inferior cerebellar artery (AICA)
Superior cerebellar artery (SCeA)
Posterior cerebral artery (PCA)
a
Supplies blood flow to spinal cord, medulla, inferior cerebellum, and
pons.
b
Supplies blood flow to brainstem, cerebellum, vermis, occipital
lobe, temporal lobe, thalamus, midbrain, and internal capsule.
Figure 26.22 Schematic illustration of the vertebrobasilar system.

The AP projection (left) demonstrates the anatomic relationship to
Table 26.14 Cervical and Intracranial Segments of the Vertebral the carotid arterial circulation (left internal carotid artery and right
Artery vertebral artery are cut away for clarity). The lateral projection (right)
provides nice delineation of the segments of the vertebral artery and
Region Segment Description their relationship to the cervical transverse foramina and foramen
Extraosseus V1 Subclavian artery to C6 transverse magnum. Abbreviations: AP, anteroposterior; AA, ascending aorta;
foramina RSCA, right subclavian artery; RICA, right internal carotid artery;
Foraminal V2 C6 to C1 transverse foramina BA, basilar artery; ACA, anterior cerebral artery; MCA, middle
Extraspinal V3 C1 to foramen magnum cerebral artery; LVA, left vertebral artery; LCCA, left common
Intradural V4 Foramen magnum to basilar artery carotid artery; LSCA, left subclavian artery; V1, extraosseous seg-
ment of VA; V2, foraminal segment of VA; V3, extraspinal segment
of VA; V4, intradural segment of VA.
anatomy. Advancement of a simple catheter or retraction of a

complex catheter and clockwise rotation will usually result in a
more stable and deeper position. If the catheter tip is coaxial
and there is no pressure clamping, DSA can be performed with and ipsilateral oblique (LAO) projections (Figs. 26.12, 26.19,
a power injector. If the position is not ideal or the catheter and 26.33). The contralateral oblique (RAO) and cranial and
recoils into the arch, the catheter can be repositioned with a caudal angulation may be useful in selected patients. The best
guidewire or exchanged for a more coaxial catheter, such as images of the intracranial circulation are AP-cranial and lateral
a modified tennis racquet, multipurpose, or glide catheter projections (Figs. 26.11, 26.19, 26.26, and 26.33); LAO caudal
(Figs. 26.27 and 26.29). In most patients, the best images of may be useful to differentiate intracranial aneurysms from
the left CCA and cervical ICA are observed in the AP, lateral, tortuous loops of intracranial arteries.
[2/7/010/8:59:48] [314–340]
Figure 26.23 Selective angiogram of a dominant left vertebral artery (VA) in the AP (left) and RAO (right) projection. Note the clear
delineation of the basilar artery, both precommunicating segments (P1) of the posterior cerebral artery (PCA), and left anterior and posterior
inferior cerebellar arteries (AICA, PICA). There is faint opacification of the left posterior communicating artery (PCOM) and postcommunicat-
ing (P2) PCA. Abbreviations: SCeA, superior cerebellar artery; BA, basilar artery; SCA, subclavian artery; IMA, internal mammary artery; V1,
extraosseous segment of VA; V2, foraminal segment of VA; V3, extraspinal segment of VA; V4, intradural segment of VA.
Table 26.15 Classification of the Posterior Cerebral Artery

Segment Region Description Branches Projection
P1 Precommunicating From BA to PCOM Perforating AP
P2 Ambient From PCOM to distal to PTA Perforating, choroid plexus, cortical AP, LAT
P3 Quadrigeminal Short medial segment Usually none AP
P4 Calcarine Termination of PCA in calcarine Cortical AP, LAT
fissure
Abbreviations: BA, basilar artery; PCOM, posterior communicating artery; PTA, posterior temporal artery; PCA, posterior cerebral artery;
AP, anteroposterior; LAT, lateral.
Selective Angiography of the Left SCA left SCA; alternatively hand or power injections can be used for
The 408 LAO projection is the best view for selective cannula- selective VA angiography using a JR or vertebral catheter. If the
tion of the left SCA (Figs. 26.1 and 26.3), and catheter selection purpose is to image the left IMA graft, the left SCA catheter can
is identical to the IA and left CCA. For complex arch config- be exchanged for an IMA or Bartorelli for selective IMA
urations, the Vitek catheter will work for virtually all patients. angiography.
Hand injections usually suffice for imaging the left SCA, and
the 308 RAO projection usually provides ideal separation of the Intracranial Circulation
VA, left IMA, and thyrocervical trunk (Figs. 26.4 and 26.5). Angiographic views of the intracranial circulation are an essen-
When it is necessary to image the VA and PCA, a power tial part of the cerebral angiogram. AP-cranial (Towne’s) pro-
injector provides the best images when the catheter is in the jections are used to characterize the ACA, MCA, PCA, circle of
[2/7/010/8:59:48] [314–340]
Figure 26.24 Schematic illustration of posterior cerebral artery (PCA)

and its cortical branches in an axial view through brainstem, temporal
lobe, and occipital lobe. Note that cortical branches arise from P2 (2) and
P4 (4), and supply major portions of the temporal and occipital lobes. 1,
Precommunicating (P1) segment of PCA; 2, postcommunicating (P2)
segment of PCA; 3, quadrigeminal (P3) segment of PCA; 4, calcarine
(P4) segment of PCA; 5, anterior temporal artery; 6, posterior temporal
artery; 7, lateral occipital artery; 8, medial occipital artery; 9, splenial
artery (cut); 10, parieto-occipital artery; 11, calcarine artery. Abbrevia-
tion: BA, basilar artery.
Table 26.16 Branches of the Posterior Cerebral Artery

Branch Segment Territory Projection
Perforating LAT
Thalamic (n ¼ 1–6)a P1 Thalamus, hypothalamus internal capsule, midbrain
Thalamogeniculate (n ¼ 2–12)a P2 (80%) Putamen, geniculate, subthalamus
P3 (20%)
Peduncular (n ¼ 0–6)a P2 Midbrain
Choroid plexus/ventricular LAT
Medial PChA P2 Choroid plexus
Lateral PChA P2 Caudate, thalamus
Cortical
Anterior temporal P2 Anterior temporal lobe AP, LAT
Posterior temporal P2 Posterior temporal lobe AP, LAT
Medial occipital P4 Parieto-occipital, corpus callosum LAT
Parieto-occipital
Calcarine
Splenial
Lateral occipital P4 Inferior temporal, occipital lobe AP, LAT
a
Numbers in parenthesis indicate the usual number of perforating branches.
Abbreviations: AP, anteroposterior; LAT, lateral; PChA, posterior choroidal artery.
[2/7/010/8:59:48] [314–340]
Figure 26.25 Selective right carotid angiogram in a patient with chronic total occlusion of the left internal carotid artery, AP-cranial (left) and
lateral (right), projections. Note extensive right to left crossover via the right anterior cerebral artery (RACA) and the anterior communicating
artery (ACOM), filling the left ACA (LACA) and middle cerebral artery (LMCA). There are two major intracranial arteries (LACA, LMCA) in the
left hemisphere, but three major arteries in the right hemisphere. In the AP-cranial projection (left), the artery (arrow) between the RACA and
right MCA (RMCA) is a persistent fetal posterior cerebral artery, which is also seen in the lateral projection (arrow) (see Figs. 26.11, 26.19,
and 26.20 for comparison).
Table 26.17 Intracranial Aneurysms Table 26.18 Sequence of Diagnostic Cerebral Angiography
Incidence Patient preparation
l General population—1% l Hydration if eGFR < 60 cc/min/1.73 m2
l Cerebral angiogram—7% l Light sedation if needed
l 2% Lidocaine for local anesthesia
Location
l Anterior circulation—90% Arterial access
Anterior communicating artery—30% l Retrograde femoral arterial approach
Middle cerebral artery—30% l 5-Fr sheath
Internal carotid artery/posterior communicating artery—30% l 5-Fr catheters
l Posterior circulation—10% l Heparin 1500–2000 IU
l Multiple—20–30% l 4-Port manifold
l Intra-arterial pressure monitoring
Rupture risk
l 6 mm—1% per year Arch study
l 7 mm—2.5% per year l 408 LAO, pigtail or TR catheter
l 25 mm (giant)—very high l Iodixanol
l Power injection (25–30 cc over 2 sec; 600 psi)
l FOV 34–42 cm
Selective cannulation
l 408 LAO guided by arch study (or tracheal air stripe)
l Simple arch: simple catheter (as described in Table 26.19) or
Willis, and their branches (Table 26.20; Figs. 26.11, 26.19, 26.20, Vitek
l Complex arch: Vitek
26.23, 26.25, 26.26, 26.32, and 26.33). It is useful to have an
understanding of intracranial collaterals and crossover, partic-
l DMA2
ularly in patients with carotid stenosis, and the location of Selective angiography
l See Tables 26.19 and 26.20
common intracranial aneurysms (Fig 26.34). Ipsilateral caudal
views are valuable for differentiating tortuous intracranial Abbreviations: eGFR, estimated glomerular filtration rate; Fr, French;
vessels from aneurysms. When using a power injector, intra- LAO. left anterior oblique; TR. tennis racquet catheter; FOV, field of
arterial contrast injection is divided into an early arterial phase view; DMA2, don’t monkey around in the arch.
[2/7/010/8:59:48] [314–340]
Figure 26.26 Selective left carotid artery angiogram using a 5-Fr Vitek catheter (lateral projection). Angiogram with hand injection (left)
resulted in streaming of contrast (arrows), and poor opacification of the stenosis at the origin of the internal carotid artery (ICA) and of the
intracranial circulation (not shown). Another angiogram was performed with a power injector (right), resulting in excellent opacification of the
ICA stenosis (arrow) and intracranial circulation.
(the artery being injected and its major branches are opacified),
a late arterial phase (reveals smaller, more distal arterial
branches), a capillary phase (resulting in a “brain blush” or
“brainogram”), and a venous phase (filling of the cerebral veins
and sinuses). By adjusting the window settings during the
capillary phase, it is possible to create a distinct brainogram.
Some patients may develop focal neurological impairment after
carotid intervention that is not associated with embolic cutoff of
a visible intracranial artery. A postintervention brainogram
may demonstrate a wedge-shaped cortical perfusion defect
consistent with distal embolization.
OTHER ANGIOGRAPHIC TECHNIQUES

Quantitative Angiography
Visual estimates are notoriously unreliable for quantitative
assessment of vessel dimensions and stenosis severity. Virtu-
Figure 26.27 Catheters commonly employed for arch aortography.
ally all angiographic systems have software that allows rapid
The modified tennis racquet (MTR) is created by amputating the digital quantitative angiography and can be useful for sizing
head of the tennis racquet, retaining a few millimeters of the radio- balloons and other interventional devices. NASCET (North
paque shaft. The MTR (or any equivalent multipurpose catheter) American Symptomatic Carotid Endarterectomy Trial) is the
may be useful for selective carotid angiography if coaxial alignment most widely used method for assessment of carotid stenosis
cannot be achieved with another simple or complex catheter. severity, using the ICA distal to the carotid bulb as the diameter
of the normal reference segment. In contrast, the ECST
[2/7/010/8:59:48] [314–340]
Table 26.19 Catheter Selection for Selective Cerebral Angiography

Catheter
configuration Catheter type Advantages Disadvantages
Simple JR, NTR, AR, VERT, Need minimal manipulation for simple Difficult to use for complex arch type
BERN, HN, BEN, JB arch type, minimal contact with
aortic wall
Complex Vitek Easy to use for simple and complex arch More contact with aortic wall than simple
types: no need to form catheter in SCA or catheter
aortic root
Complex SIM Useful for complex arch configurations or if More contact with aortic wall; need to form
Vitek fails catheter in SCA or aortic root
Multipurpose MPa, Glide, MTRa,b Excellent coaxial alignment for virtually all Cannot be used to cannulate the great
vessel configurations and arch types vessels; requires guidewire and catheter
exchange
Producer names are given in the text.
a
Catheter with side holes; all others are end-hole catheters; PSI ¼ 500–550 psi for side-hole catheters, 400–450 psi for end-hole catheters.
b
MTR is formed by amputating the head of the tennis racquet catheter, leaving 1–2 mm of the radiopaque shaft in place (Fig. 26.27).
Abbreviations: JR, Judkins right; NTR, no torque right; AR, Amplatz right; VERT, vertebral; BERN, Bernstein; HN, Headhunter; BEN,
Bentson; JB, jugular bulb; SIM, Simmons; MP, multipurpose; MTR, modified tennis racquet; SCA, subclavian artery.
Table 26.20 Technique of Selective Cerebral Angiography

Catheter Target Injection
position vessel technique Projection Catheter
IA IA Hand RAO JR, Vitek
SCA R-SCA Hand LAO, RAO JR, Vitek
R-VA Powera AP, RAO, JR, Vitek
LAT
R-PCA Powera AP, LAT JR, Vitek
R-VA R-VA Hand/ AP, RAO, JR, Vertebral
Powerb LAT
R-PCA Hand/ AP-C, LAT JR, Vertebral
Powerb
R-CCA R-CCA Powerc AP, RAO JR, Vitek, MPd
R-ICA Powerc AP, RAO, JR, Vitek, MPd
LAT
R-ACA/ Powerc AP-C, LAT JR, Vitek, MPd
MCA
L-CCA L-CCA Powerc AP, LAO JR, Vitek, MPd
L-ICA Powerc AP, LAO, JR, Vitek, MPd
LAT
L-ACA, Powerc AP-C, LAT JR, Vitek, MPd
MCA
L-SCA L-SCA Hand LAO, RAO JR, Vitek
L-VA Powera AP, LAO, JR, Vitek
LAT
L-PCA Powera AP, LAT JR, Vitek
L-VA L-VA Hand/ AP, LAO, JR, Vertebral Figure 26.28 Relationship of the tracheal air stripe (dotted lines) to
Powerb LAT the origin of the great vessels. In the 408 LAO projection, the
L-PCA Hand/ AP-C, LAT JR, Vertebral tracheal air stripe is a good landmark for selective cannulation of
Powerb the innominate artery (proximal to the air stripe), left carotid artery
a (middle of the air stripe), and left subclavian artery (distal to the air
8–10 total volume, 4–6 cc/sec, 450–500 psi.
b stripe). Abbreviation: LAO, left anterior oblique.
5–7 cc total volume, 3–5 cc/sec, 450–500 psi.
c
8–12 cc total volume, 6–9 cc/sec, 450–500 psi (end hole), 500 psi
(side hole).
d
Any multipurpose catheter, such as MP, Glide catheter, or MTR.
Abbreviations: IA, innominate artery; SCA, subclavian artery; VA,
vertebral artery; PCA, posterior cerebral artery; CCA, common
carotid artery; ICA, internal carotid artery; MCA, middle cerebral
(European Carotid Surgery Trial) method uses the carotid
artery; ACA, anterior cerebral artery; R, right; L, left; RAO, right
bulb as the normal reference diameter, and consistently
anterior oblique; LAO, left anterior oblique; LAT, lateral; AP,
anteroposterior; AP-C, anteroposterior with cranial angulation; JR, yields higher percent stenoses than the NASCET method
Judkins right; MP, multipurpose; MTR, modified tennis racquet. (Fig. 26.35) (9).
[2/7/010/8:59:48] [314–340]
Figure 26.29 Catheters commonly employed for selective cannulation and angiography of the great vessels (see text for details). Simple
catheters (left) have a simple primary curve and minimal secondary curve (from left to right: Headhunter, JB-1, JR-4, MPA, vertebral). Note that
the multipurpose (MPA), glide catheter (not shown), and modified tennis racquet (see Fig. 26.27) are not readily suited for selective cannulation,
but are useful for selective angiography when other catheters do not provide coaxial alignment. Complex catheters (right) have more complex
relationships between the catheter tip and shaft (left ¼ Simmons, right ¼ Vitek).
Figure 26.30 Schematic illustration of manipulation of simple and complex catheters for selective angiography of great vessels. (A) Simple
catheter manipulation involves placement of the catheter in the ascending aorta, followed by retraction and counterclockwise rotation to
engage the innominate artery (IA). Retraction and counterclockwise rotation are repeated to engage the left common (CCA) and left
subclavian arteries (SCA). (B) Vitek catheter manipulation involves positioning the catheter distal to the left SCA, followed by advancement
and counterclockwise rotation into the left SCA. While retaining the curve of the catheter, the Vitek is advanced to the left CCA and IA.
(C) Simmons catheter manipulation involves positioning the catheter in the ascending aorta (or left SCA), and forming a curve in the catheter
by advancing it over a J-wire. After removing the J-wire, the Simmons can be retracted and rotated counterclockwise to engage the left SCA.
While retaining the curve, the Simmons is advanced to the left CCA and IA, similar to the Vitek.
[2/7/010/8:59:48] [314–340]
Figure 26.31 Angiography of the innominate artery (IA) and its bifurcation (dotted box) into right subclavian (SCA) and common carotid
arteries (CCA). The arch aortogram (408 LAO) demonstrates a type 1 arch with the usual configuration of the great vessels (left). This
projection obscures the IA bifurcation into the right SCA and CCA, and results in close overlap of the right vertebral artery (VA) and CCA.
Selective angiography of the IA (308 RAO) results in better separation of the right SCA and CCA (right), and no overlap of the right VA and
CCA. Note the “string sign” in the right internal carotid artery (arrows). Abbreviations: LAO, left anterior oblique; RAO, right anterior oblique.
Figure 26.32 Selective right carotid artery angiography in a patient with asymptomatic carotid stenosis. Lateral projection demonstrates a
focal, ulcerated, severe stenosis at the origin of the internal carotid artery (ICA). The intracranial circulation is normal in the AP (middle) and
lateral (right) projections. Note faint filling of the right vertebral artery from reflux of contrast during power injection (arrow).
Gradient Assessment translesional pressure gradients are useful for assessment of

and Intravascular Ultrasound SCA stenoses but are rarely necessary (and may increase the
Measurement of translesional pressure gradient is useful for risk of stroke) in patients with ICA stenosis. In addition, the
assessing the functional significance of stenoses and may be established benefit of carotid revascularization in terms of
measured at rest and after intra-arterial nitroglycerin (200–400 mg). stroke prevention has been based on stenosis severity and not
A peak-to-peak pressure gradient 20 mmHg is considered translesional gradient. Intravascular ultrasound is feasible in the
hemodynamically significant. From a practical standpoint, brachiocephalic circulation but has not been widely adopted.
[2/7/010/8:59:48] [314–340]
Figure 26.33 Selective left carotid artery angiogram in an elderly patient with symptomatic left carotid stenosis. There is eccentric severe
stenosis of the origin of the left internal carotid artery (ICA) at the level of the carotid bulb (CB) and multiple sequential moderate stenosis
(arrows) in the petrous segment of the ICA (left, 408 LAO projection). Images of the intracranial circulation in the AP-cranial (middle) and
lateral (right) projection demonstrate normal antegrade filling of the left anterior (ACA) and middle (MCA) cerebral arteries. Note left to right
crossover via the anterior communicating artery (arrowhead), associated with asymptomatic severe stenosis of the right ICA (not shown), and
considerable atherosclerosis in the petrous and cavernous left CCA (arrows).
Figure 26.34 Schematic illustration of the circle of Willis demonstrating

typical locations of intracranial aneurysms (see Table 26.20). Abbrevia-
tions: ACOM, anterior communicating artery; A1, precommunicating
anterior cerebral artery (ACA); A2, postcommunicating ACA; ICA, internal
carotid artery; MCA, middle cerebral artery; PCOM, posterior communi-
cating artery; P1, precommunicating posterior cerebral artery (PCA); P2,
postcommunicating PCA; BA, basilar artery; SCeA, superior cerebellar
artery; AICA, anterior inferior cerebellar artery; PICA, posterior inferior
cerebellar artery; VA, vertebral artery.
COMPLICATIONS site complications (hematoma, dissection, AV fistula, pseudoa-

In addition to complications associated with other invasive neurysm, hemorrhage) and systemic complications (contrast
angiographic procedures, cerebral angiography has unique reactions, contrast nephropathy, cholesterol embolization) are
complications related to the central nervous system. Access encountered in <5% of patients undergoing invasive
[2/7/010/8:59:48] [314–340]
Figure 26.35 Methodology for quantitative carotid artery angiography. Calculations for percent diameter stenosis vary depending on the site
of the normal reference segment. The NASCET method utilizes the proximal internal carotid artery (C) as the reference segment, whereas the
ECST method utilizes the carotid bulb (B) as the reference segment. Abbreviations: NASCET, North American Symptomatic Carotid
Endarterectomy Trial; ECST, European Carotid Surgery Trial.
angiographic procedures, including cerebral angiography.

Neurological complications are the most feared complications
of cerebral angiography. Causes of neurological events during
cerebral angiography include embolism (plaque, thrombus, air)
and direct arterial injury (vasospasm, dissection, perforation)
(Fig. 26.36). Although older series reported death, stroke, and
transient ischemic attacks (TIAs) in 0.06%, 0.14%, and 2% of
patients undergoing cerebral angiography respectively (10), the
incidence of neurologic complications appears to have
decreased substantially. In a recent series from 2001 to 2006,
procedure-related TIAs occurred in 0.34%, none resulting in
permanent neurologic deficit (11). Factors accounting for the
decrease in complications have not been precisely defined, but
may include the use of intra-arterial pressure monitoring,
heparin anticoagulation, isosmolar contrast, and better pre-
formed catheter designs requiring less catheter manipulation.
The potential for complications mandates that neurologic
surveillance be performed during and after the procedure.
Light conscious sedation is reasonable, but deep sedation
should be avoided. Angiography and recovery room staff
should be trained to evaluate changes in consciousness, cogni-
tive and language deficits, visual complaints, headache, gait
and vestibular symptoms, and focal motor and sensory deficits. Figure 26.36 Catheter-induced dissection (arrowhead) of the left
Physicians performing cerebral angiography must be compe- common carotid artery (558 RAO). Severe asymptomatic eccentric
tent in performing a thorough neurological examination, dissection (left) was treated by successful direct carotid stenting
be familiar with National Institutes of Health Stroke Scale using embolic protection (right). Abbreviation: RAO, right anterior
oblique.
(NIHSS) (12,13), and recognize criteria for immediate stroke
intervention.
[2/7/010/8:59:48] [314–340]
REFERENCES 8. Hayashi N, Sakai T, Kitagawa M, et al. Radiation exposure to

1. Collins P, McKay I, Rajagoplan S, et al. Is carotid duplex scanning interventional radiologists during manual-injection digital sub-
sufficient as the sole investigation prior to carotid endarterec- traction angiography. Cardiovasc Interv Radiol 1998; 21:240–243.
tomy? Br J Radiol 2005; 78:1034–1037. 9. Osborn AG, ed. Diagnostic Cerebral Angiography. 2nd ed. Phil-
2. Khaw KT. Does carotid duplex imaging render angiography redun- adelphia, PA: Lippincott Williams & Wilkins, 1999:373.
dant before carotid endarterectomy? Br J Radiol 1997; 70:235–238. 10. Kaufmann TJ, Huston J III, Mandrekar JN, et al. Complications of
3. Casserly IP, Yadav JS. Carotid Intervention. In: Casserly IP, Sachar diagnostic cerebral angiography: evaluation of 19,826 consecutive
R, Yadav JS, eds. Manual of Peripheral Vascular Intervention. patients. Radiol 2007; 243:812–819.
Philadelphia, PA: Lippincott Williams & Wilkins, 2005:83–109. 11. Dawkins AA, Evans AL, Wattam J, et al. Complications of cerebral
4. Patel P, Shammas NW, Kapalis MJ, et al. Routine visualization of angiography: a prospective analysis of 2,924 consecutive proce-
the left internal mammary artery before bypass surgery: is it dures. Neuroradiol 2007; 49:753–759.
necessary? J Invasive Cardiol 2005; 17:479–481. 12. Howard G, Shelton BJ, Howard VJ. Statistical and experimental
5. Matsumoto Y, Hongo K, Toriyama T, et al. Transradial approach design issues in the evaluation of carotid artery stenting. In:
for diagnostic selective cerebral angiography: results of a consec- Al-Mubarak N, Roubin GS, Iyer SS, et al., eds. Carotid Artery
utive series of 166 cases. AJNR Am J Neuroradiol 2001; 22:704–708. Stenting: Current Practice and Technique. Philadelphia, PA:
6. Layton KF, Kallmes DF, Kaufman N. Use of the ulnar artery as an Lippincott Williams & Wilkins, 2004:313–326.
alternative access site for cerebral angiography. AJNR Am J 13. Stroke Scales and Clinical Assessment Tools. NIH Stroke Scale
Neuroradiol 2006; 27:2073–2074. (NIHSS). Available at: http://www.strokecenter.org/trials/scales/
7. Sievert H, Ensslen R, Fach A, et al. Brachial artery approach for nihss.html.
transluminal angioplasty of the internal carotid artery. Cathet
[2/7/010/20:47:53] [341–354]
27
Application of intracoronary physiology: use of pressure

and flow measurements in clinical practice
Olivier Muller and Bernard De Bruyne
INTRODUCTION relation will be important when considering a stenosis in a

The goal of any treatment is to improve patient prognosis and/ large artery [i.e., left anterior descending (LAD)] compared
or symptoms. Accordingly, the goal of any diagnostic tool is to with a smaller one.
guide decision making to apply optimal treatment in individual
patients. Any diagnostic tool not fulfilling these requirements
should not be used in patients. Keeping this in mind, this
Fractal Nature of the Coronary Circulation
The arterial tree is by nature made of arterial bifurcations and
chapter will briefly review what is often referred to as “invasive
follows a stepwise reduction of vascular diameter down to the
coronary physiology.”
capillary level. Each bifurcation consists of an asymmetric
As opposed to anatomical (also referred to as “morpho-
reduction according to the law of conservation of energy. In
logical”) approaches, physiological (“functional”) measure-
fact, the coronary tree has been shown to be governed by fractal
ments assess the function of the coronary circulation. The
geometry, in other words, at whatever level the coronary tree is
function of the coronary arterial circulation is to provide
observed, the morphology of the tree is similar. In fractal
nutrients to the myofibrils. Accessorily, the coronary arteries
models, the quantity to be scaled is related to the power of a
play a role in diastolic function (turgor effect). No other organ
fraction. A recent study by Finet et al. proposed a simple and
is so dependent on the function of the organ it perfuses.
accurate fractal ratio between the diameters of the mother and
The combination of an accurate anatomic assessment and
the two daughter vessels (3). This ratio—diameter of the
precise functional information is indispensable to tailor the
mother vessel ¼ 0.678 (diameter of the first daughter þ
treatment of patients with suspected or known coronary artery
diameter of the second daughter vessel)—confirmed other
disease.
studies and allows appreciation of the luminal diameter of
each daughter vessel (4,5).
ANATOMICAL CONSIDERATIONS
Three-Layer Histology of the Coronary Arteries PHYSIOLOGICAL CONSIDERATIONS
Normal arteries have a well-developed trilaminar structure— A detailed review of myocardial flow regulation is beyond the
the intima, media and the adventitia. The endothelial cell of the scope of this chapter. Yet, a reminder of a number of aspects is
tunica intima constitutes the barrier with blood and plays an useful to understand the basics of the physiological assessment
important role in the regulation of hemostasis and vascular of the coronary circulation.
tone. The strategic location of the endothelium allows it to
sense changes in hemodynamic forces and blood-borne signals
and to respond by releasing vasoactive substances. A balance A Few Basic Concepts
between endothelium-derived relaxing (i.e., nitric oxide) and Flow, Pressure, and Resistance
contracting factors (i.e., endothelin) maintains vascular homeo- The main parameters of the circulatory function are flow,
stasis. When this balance is disrupted (i.e., atherosclerosis), it pressure and resistance.
predisposes the vasculature to vasoconstriction and thus, to Q ¼ DP=R
disturbance in coronary blood flow (1).
Measuring absolute coronary blood flow (in mL/min) and
absolute resistance (in mmHg/mL/sec) sounds like the
The Size of the Arteries Is Proportional “alpha and omega” of vascular physiology. However, even if
to Myocardial Mass absolute flow and resistance were to be measured easily and
In normal individuals, there is a close correlation between the reliably in patients, it is most likely that their impact on clinical
lumen cross-sectional area of a coronary artery at each point decision making would be rather modest. Flow—and therefore
along its length and the corresponding regional myocardial resistance—both depend on the myocardial mass to be per-
mass. In contrast, in patients with coronary atherosclerosis, fused. Therefore, there is no unequivocal normal value for
measured coronary artery lumen area is diffusely 30% to 50% coronary flow and resistance. In contrast, under normal con-
too small for distal myocardial bed size compared with normal ditions, coronary pressure equals central aortic pressure over
subjects (2). Figure 27.1 shows the relation between the size of a the entire length of the epicardial arteries, even during hyper-
given artery and the mass of myocardium perfused by this emia. This unique characteristic of coronary pressure gives the
artery. In patients with coronary artery disease, the dimension interventionalist an unequivocal reference value: whatever
of a given artery is too small for the mass of myocardium. This the myocardial mass, the size of the artery, the systemic
[2/7/010/20:47:53] [341–354]
Figure 27.1 Relation between the size of epicardial coronary

artery and myocardial mass in normal patients and in patients with
coronary artery disease. Source: From Ref. 2.
hemodynamics, the age of the patient, the status of the micro-

vasculature, etc., the pressure in the distal part of an epicardial Figure 27.2 Plot of the relation between reductions in endocardial
artery should be identical to central aortic pressure. If this is not flow and wall thickening reductions. Source: From Ref. 6.
the case (i.e., in case of a pressure gradient), it necessarily
implies that the resistance of the epicardial artery is abnormal.
Coronary Flow and Myocardial Contractions

Because of the squeezing effect of the contracting ventricles,
coronary blood flow occurs mainly during diastole. This dia-
stolic preeminence of coronary blood flow is most pronounced
at rest (as compared with hyperemia) and in the left coronary
artery. In case of epicardial stenosis, the pressure gradient will
therefore be mainly diastolic (“no flow, no gradient”).
Flow-Function Relationship
Myocardial blood flow represents approximately 5% of cardiac
output. Because of its constant work, resting myocardial oxy-
gen demand is high and the extraction of oxygen by the
myocardium is close to its maximum, much higher than in
other organs. Oxygen saturation of the coronary sinus venous
blood is close to 20% (as a comparison, the oxygen saturation in
the renal vein is 85%). Hence, as extraction cannot increase
much further, the coronary circulation can only meet oxygen
demand by fine tuning myocardial blood flow. In addition, as
soon as myocardial flow decreases below 90% of its normal
resting levels, myocardial function starts to decrease (Fig. 27.2). Figure 27.3 Since the epicardial vessels contribute only a minimal
It is obvious that the control of myocardial blood flow must be fraction of the total vascular resistance there is no significant pres-
remarkably tight to avoid wall motion abnormalities. Con- sure drop along the conductance vessels. In contrast, passage
versely, this also implies that when myocardial wall motion is through the resistive vessels produces a large drop in pressure.
normal, its resting perfusion must be normal. Abbreviations: FFR, fractional flow reserve; CFR, coronary flow
reserve; IMR, microvascular resistance.
The Control of Myocardial Flow

The regulation of myocardial blood flow is multifactorial. The
neuro-humoral, endothelial, endo- and paracrine, metabolic, and vessels” because they do not oppose any resistance to blood
physical factors are largely nonlinear, cumulative and interacting. flow. The second compartment consists of arteries <400 mm, or
It is therefore very difficult to study these factors individually. “resistive vessels” (Fig. 27.3). When no stenosis is present,
To simplify, the coronary circulation can be considered a myocardial flow is primarily controlled by resistive vessels
two-compartment model. The first compartment consists of (7), as they are able to vasodilate under physiological and
epicardial vessels, which are also referred to as “conductance pharmacological stress. At coronary angiography they are not
[2/7/010/20:47:53] [341–354]
APPLICATION OF INTRACORONARY PHYSIOLOGY 343
clearly delineated but appear as a myocardial blush of contrast concept of index of microvascular resistance (IMR) considering
medium. During exercise or any other form of increased oxy- that the mean transit time during maximal hyperemia is
gen demand, the resistance of the microvasculature decreases inversely proportional to hyperemic flow. Therefore, during
allowing for an increased blood flow. Similarly, when a stenosis maximal hyperemia:
is present in the epicardial artery, this increased epicardial
IMR ¼ Pd =1=Tmn ¼ Pd Tmn
resistance is compensated by an equivalent decrease in micro-
vascular resistance. This results in a maintained total resistance where Pd is the distal coronary pressure and Tmn is the mean
to blood flow and a preserved resting flow, with residual— transit time. IMR is specific for the microcirculation and is
albeit reduced—coronary flow reserve (CFR). When the epi- simple to obtain as Pd and Tmn can be obtained simultaneously
cardial stenosis progresses further, its relative contribution to with the PressureWire1 (St. Jude Medical, St. Paul, Minnesota,
total resistance increases. At the extreme, when the stenosis U.S./Radi Medical Systems, Uppsala, Sweden). It has been well
becomes “critical,” the compensation capacity of the micro- validated in animals (10) and was recently used in the setting of
vascular circulation is exhausted. Any additional increase in acute coronary syndromes to predict clinical outcomes (11) and
epicardial resistance will result in an increase in total resistance assess the effect of treatment (12).
and in a decrease in myocardial flow.
Hyperemic Stenosis Resistance
PHYSIOLOGICAL INDICES The resistance of an epicardial stenosis can be given by the ratio
OF THE CORONARY CIRCULATION of the pressure gradient and the flow across that particular
Since the seminal work of L.K. Gould (8), several indices of stenosis. Accordingly, Meuwissen et al. introduced the concept
coronary physiology have been proposed to guide clinical of hyperemic stenosis resistance (HSR), which can be calculated
decision making. Fractional flow reserve (FFR) is the best as the ratio of transstenotic gradient (Pa – Pd) to average peak
validated index. Therefore, we will briefly describe the other flow velocity (Vel) during maximal hyperemia.
indices then we will focus on FFR. Pa Pd
HSR ¼
Vel
Coronary Flow Reserve This measurement requires both a flow and a pressure sensor
CFR is defined as the ratio of hyperemic blood flow (Qmax) to (ComboWire, Volcano, Inc., Rancho Cordova, California, U.S.)
resting myocardial blood flow (Qrest). and is specific for the epicardial stenosis (13,14).
Qmax
CFR ¼
Qrest
FRACTIONAL FLOW RESERVE
Normal CFR value is between 4 and 6, which means that Definition
microvascular resistance can decrease by a factor of 4 to 6. FFR is the ratio of maximal myocardial blood flow depending
Since absolute myocardial flow is not easy to determine, surro- on a stenotic artery to maximal myocardial blood flow if that
gates of flow are commonly used: for example, flow velocities same artery were to be normal. In other words, it is a fraction of
assessed by the Doppler Wire (FloWire, Volcano, Inc., Rancho the maximal normal flow assuming that these measurements
Cordova, California, U.S.) or mean transit time (Tmn) assessed are obtained when the microvasculature resistance is minimal
by the PressureWire1 (St. Jude Medical, St. Paul, Minnesota, and constant (maximal hyperemia) (15,16).
U.S./Radi Medical Systems, Uppsala, Sweden). Regardless of FFR represents the extent to which maximal myocardial
technical/practical aspects, the concept of CFR has several blood flow is limited by the presence of an epicardial stenosis.
limitations: (i) resting flow, which appears as the denominator, If FFR is 0.60, it means that maximal myocardial blood flow
is highly variable; (ii) hyperemic flow is directly dependent on reaches only 60% of its normal value. Conversely, FFR provides
systemic blood pressure; (iii) the hyperemic and resting meas- the interventionalist with the exact extent to which optimal
urements are not performed simultaneously but successively; stenting of the epicardial stenosis will increase maximal myo-
and (iv) CFR is not specific for the epicardial stenosis. Stated cardial blood flow. An FFR of 0.60 implies that stenting the
another way, when CFR is too low, it is impossible to tell focal stenosis responsible for this abnormal FFR should bring
whether this abnormal value is related to a stenosis in the FFR to 1.0, which represents an increase in maximal myocardial
epicardial artery, to microvascular disease, or to a combination blood flow of 67%.
of both. For these reasons, CFR is of limited value for clinical
decision making (9).
Calculation
FFR is a ratio of two flows. It has been shown, however, that
The Index of Microvascular Resistance this ratio of two flows can be derived from two pressures (distal
The resistance of a vascular system is given by the ratio of the coronary pressure and aortic pressure), provided they are both
pressure gradient to the flow across that particular system. measured during maximal hyperemia. The theoretical explana-
Accordingly, the resistance (R) of the coronary microvascular tion of this relationship between hyperemic flows and hypere-
compartment is the ratio of mic pressures is displayed in Figure 27.4.
Pd Pv
R¼
Q
Equipment
where Pd is distal coronary arterial pressure and Pv coronary Catheters
venous pressure, or right atrial pressure. In the coronary circulation The use of diagnostic catheters is technically feasible (17). Yet,
Pv is often almost negligible. Fearon et al. introduced the because of higher levels of friction hampering wire manipulation,
[2/7/010/20:47:53] [341–354]
Figure 27.4 (A) Simplified theoretical explanation illustrating how a ratio of two flows can be derived from a ratio of two pressures provided
these pressures are recorded during maximal hyperemia. (B) Concept of FFR measurements. When no epicardial stenosis is present (black
lines) the driving pressure Pa determines a normal (100%) maximal myocardial blood flow. In case of stenosis responsible for a hyperemic
pressure gradient of 30 mmHg (grey lines), the driving pressure will no longer be 100 mmHg but 70 mmHg (Pd). Since during maximal
hyperemia the relationship between driving pressure and myocardial blood flow is linear, myocardial blood flow will only reach 70% of its
max max
normal value. This numerical example shows how a ratio of two pressure (Pd /Pa) corresponds to a ratio of two flows QS =QN . It also
illustrates how important it is to induce maximal hyperemia. Abbreviation: FFR, fractional flow reserve.
the smaller internal caliber prejudicing pressure measurements, U.S.). The sensor is located 3 cm from the tip, at the junction
and the inability to perform ad hoc percutaneous coronary between the radio-opaque and the radiolucent portions. The
intervention (PCI) using diagnostic catheters, the use of guiding last generations of these 0.014-in wires have similar handling
catheters is highly recommended. characteristics to most standard angioplasty guide wires.
Wires Hyperemia
Measuring intracoronary pressure requires the use of a specific To measure FFR, it is absolutely essential to achieve maximal
solid state sensor mounted on a floppy-tipped guide wire. In vasodilatation of the two vascular compartments of the coro-
mainstream practice two such systems are available at this nary circulation, namely the conductance arteries (epicardial)
time, namely the PressureWire1 (St. Jude Medical, St. Paul, and the resistance arteries (microvasculature). The pharmaco-
Minnesota, U.S./Radi Medical Systems, Uppsala, Sweden) and logical agents most often used to induce hyperemia are listed in
the WaveWire1 (Volcano, Inc., Rancho Cordova, California, Table 27.1 (18–20).
[2/7/010/20:47:53] [341–354]
Table 27.1 Importance of Epicardial and Microvascular Vasodilatation When Measuring Fractional Flow Reserve
Plateau (sec) Half-life (min) Dose
Epicardial vasodilation
Isosorbide dinitrate – – At least 200 mg IC bolus, at least 30 sec before the first measurements
Microvascular vasodilation
Adenosine IC 5–10 0.5–1 50 mg IC bolus
Adenosine IV 60–120 1–2 140 mg/kg/mina
Papaverine IC 30–60 2 8 mg in the right coronary artery, 12 mg in the left coronary artery
Nitroprusside IC 20 1 0.6 mg/kg in bolus
a
preferably through a central venous (e.g., femoral) line.
The epicardial arteries A bolus of 200 mg isosorbide Anticoagulation

dinitrate (or any other form of intracoronary nitrates) allows As soon as a device is advanced into the coronary tree, the use
the abolition of any form of epicardial vasoconstriction. It is of the same anticoagulation regimens as employed during a
good clinical practice to give intracoronary nitrates when PCI procedure is recommended: heparin adjusted to weight,
performing coronary angiograms, especially when a wire is validated by a monitored activated coagulation time (ACT) of
manipulated in the coronary tree. In addition, it is important to at least 250 seconds. Patients are supposed to be under aspirin
realize that all the data on FFR and its relation with noninvasive treatment.
stress modalities or with clinical outcome have been obtained
after intracoronary administration of nitrates.
Microvascular circulation Inducing maximal hyperemia Unique Characteristics of FFR
to obtain physiological information about a stenosis might be FFR has a number of unique characteristics that make this
compared with placing the stenosis in a wind tunnel. From a index particularly suitable for functional assessment of coro-
theoretical point of view, the concept of FFR is based on nary stenoses and clinical decision making in the catheter-
achieving minimal microvascular resistance (Fig. 27.4). There- ization laboratory.
fore, expressions like “baseline FFR” or “FFR at rest” are phys-
iological nonsense. This also extends to all other circulations. “Universal” Normal Value Is 1
Even when the resting pressure gradient is large we recommend An unequivocally normal value is easy to refer to but is rare in
the induction of hyperemia because it allows us to evaluate what clinical medicine. Since in a normal epicardial artery there is
is the residual resistance reserve. virtually no decline in pressure, not even during maximal
An example of a typical coronary pressure tracing during the hyperemia, it is obvious that Pd/Pa will equal or be very
administration of intravenous adenosine is shown in Figure 27.5. close to unity. FFR was obtained in 37 arteries in 10 individuals
Figure 27.5 (See color insert ) Typical example of simultaneous aortic pressure (Pa) and distal coronary pressure (Pd) recordings at rest and
during maximal steady-state hyperemia as induced by an intravenous infusion of adenosine. Soon after starting the infusion, the decrease in
distal pressure is preceded by a transient increase in aortic pressure. Abbreviation: FFR, fractional flow reserve.
[2/7/010/20:47:53] [341–354]
Table 27.2 Cutoff Values for Fractional Flow Reserve

Authors Reference Patients Number Test Threshold
De Bruyne et al. 22 1 VD 60 Bicycle ECG 0.72a
Pijls et al. 21 1 VD before and post percutaneous 60 Bicycle ECG 0.74a
coronary intervention
Pijls et al. 23 1 VD, intermediate stenosis 45 Bicycle ECG, thallium þ 0.75a
dobutamine echo
Bartunek et al. 24 1 VD 75 Dobutamine echo 0.78a
Chamuleau et al. 25 Multivessel disease 127 MIBI-Spect 0.74b
Abe et al. 26 1 VD 46 Thallium 0.75a
De Bruyne et al. 7 Post myocardial infarction 57 MIBI-Spect 0.75–0.80a
a
100% specificity.
b
optimal cutoff value.
Abbreviation: VD, vessel disease.
without atherosclerosis (group I) and in 106 nonstenotic arteries reproducible. In addition FFR has been shown to be independent
in 62 patients with arteriographic stenoses in another coronary of gender and risk factors such as hypertension and diabetes
artery (group II). In group I, the FFR was near unity (0.97 0.02; (28). These characteristics contribute to the accuracy of the
range 0.92–1), indicating no resistance to flow in truly normal method and to the trust in its value for decision making.
coronary arteries, but it was significantly lower (0.89 0.08;
range 0.69–1) in group II, indicating a higher resistance to Takes into Account the Contribution of Collaterals
flow (7). Whether myocardial flow is provided antegradely by the
Thus, it is important to realize that in normal-looking epicardial artery or retrogradely through collaterals does not
coronary arteries in patients with proven atherosclerosis else- really matter for the myocardium. Distal coronary pressure
where, the epicardial coronary arteries may contribute to total during maximal hyperemia reflects both antegrade and retro-
resistance to coronary blood flow even though there is no grade flow according to their respective contribution. This
discrete stenosis visible on the angiogram. In approximately holds for the stenoses supplied by collaterals but also for
50% of these arteries, FFR is lower than the lowest value found stenosed arteries providing collaterals to another more criti-
in strictly normal individuals. In approximately 10% of athero- cally diseased vessel (Fig. 27.6).
sclerotic arteries, FFR will be lower than the ischemic threshold.
Practically speaking, this finding implies that myocardial ische- Specifically Relates Severity of the Stenosis
mia might be present in atherosclerotic patients in the absence to Mass of Tissue to Be Perfused
of discrete stenoses (7,21). The larger the myocardial mass subtended by a vessel, the
larger the hyperemic flow, and in turn, the larger the gradient
Has a Well-Defined Cutoff Value and the lower the FFR. This explains why a stenosis with a
Cutoff or threshold values are values that distinguish normal minimal cross-sectional area of 4 mm2 has totally different
from abnormal levels for a given measurement. To enable hemodynamic significance in the proximal LAD artery versus
adequate clinical decision making in individual patients it is the second marginal branch. It also explains FFR measurements
paramount that any level of uncertainty is reduced to a mini- of collaterals, where the mass supplied by an artery can change
mum. Cutoff value for FFR has been evaluated in several after revascularization of the retrograde supplied vessel.
studies and compared with several decision-making modal-
ities, among them radionuclide perfusion was the most used Has Unequaled Spatial Resolution
(9). Stenoses with FFR measurement of <0.75 are almost The exact position of the sensor in the coronary tree can be
invariably able to induce myocardial ischemia (cutoff with a monitored under fluoroscopy, and documented angiographi-
specificity of 100%, a sensitivity of 88%, a positive predictive cally. Pulling back the sensor under maximal hyperemia (usu-
value of 100% and an overall accuracy of 93%). Stenoses with ally adenosine IV) provides the operator an instantaneous
an FFR > 0.80 are almost never associated with exercise assessment of the abnormal resistance of the arterial segment
induced ischemia (Table 27.2). This means that the “grey located between the guide catheter and the sensor. While other
zone” for FFR (0.75–0.80) spans over 6% to 7% of the entire functional tests reach a “per patient” accuracy (exercise ECG),
range of FFR values. or at best a “per vessel” accuracy (myocardial perfusion imag-
ing), FFR reaches a “per segment” accuracy with a spatial
Is Not Influenced by Systemic Hemodynamics resolution of a few millimeters.
In the catheterization laboratory systemic pressure, heart rate
and left ventricular contractility are prone to change. In contrast
to many other indices measured in the catheterization labora- Clinical Applications
tory, changes in systemic hemodynamics do not influence the Angiographically Dubious Stenoses
value of FFR in a given coronary stenosis (27). This is due not The main general indication for FFR is the precise assessment of
only to the fact that aortic and distal coronary pressures are the functional consequences of a given coronary stenosis with
measured simultaneously, but also to the extraordinary capa- unclear hemodynamic significance (23). Cardiologists describe
bility of the microvasculature to repeatedly vasodilate to exactly angiographic coronary narrowings with uncertain functional
the same extent. In addition, FFR measurements are extremely consequences as mild-to-moderate stenoses, dubious lesions,
[2/7/010/20:47:53] [341–354]
Figure 27.6 (See color insert ) Example of the influence of collaterals on FFR measurements in a 76-year-old man with a critical stenosis in
the proximal RCA (A) and collaterals supplied by the left coronary artery (B). The FFR in the distal LAD was measured first (A and D) before
recanalization of the RCA and (C and E) after recanalization of the RCA. When antegrade flow was restored in the RCA, the LAD had no
longer to supply blood to the territory of the RCA. Therefore, hyperemic flow in the LAD was lower than before and the FFR increased from
0.76 to 0.82. This example also illustrates the relationship between FFR and the myocardial mass supplied by the artery: the larger the
myocardial mass, the greater the hyperemic flow, and the lower the FFR for a given stenosis. Abbreviations: FFR, fractional flow reserve;
RCA, right coronary artery; LAD, left anterior descending artery.
intermediate stenoses, non–flow limiting stenoses, etc. Angio- these noninvasive tests are often contradictory, which renders
graphic assessment of such stenoses produces insufficient infor- appropriate clinical decision making difficult (Fig. 27.7). In
mation to determine whether a stenosis is hemodynamically addition, the clinical outcome of patients in whom PCI has
significant or not. Moreover, angiographic assessment is often been deferred, because the FFR indicated no hemodynamically
the only decision-making modality to perform an angioplasty. significant stenosis, is very favorable. In this population the risk
Even if the clinical cardiologist is offered many options and of death or myocardial infarction is approximately 1% per year,
techniques for noninvasive functional evaluation, it has been and this risk is not decreased by PCI (31). These results strongly
report that up to 71% of PCIs are being performed in the support the use of FFR measurements as a guide for decision
absence of any sort of functional evaluation (29). This scenario, making about the need for revascularization in “intermediate”
often referred to as the “oculostenotic reflex,” is even more lesions. Figure 27.8 illustrates how two angiographically similar
worrisome now that safety concerns associated with late stent stenoses may have a completely different hemodynamic sever-
thrombosis have been identified (30). ity. One of them should be revascularized, the other not. Based
Direct translesional pressure measurements correlate solely on the angiogram, the decision should be identical in
well with noninvasive assessment of coronary artery disease. both cases, which would lead to an inappropriate interven-
In a study of 45 patients with angiographically dubious sten- tional decision in one of these patients.
oses it was shown that FFR has a much larger accuracy in
distinguishing hemodynamically significant stenoses than exer- Left Main Stem Disease
cise ECG, myocardial perfusion scintigraphy and stress echo- The presence of a significant stenosis in the left main stem is of
cardiography taken separately (23). Furthermore, the results of critical prognostic importance (32). Conversely, revascularization
[2/7/010/20:47:53] [341–354]
Figure 27.7 Plots of the FFR in 45 patients with an

angiographically intermediate stenosis according to the
results of noninvasive testing. The hollow circles repre-
sent negative tests. The black dots represent positive
tests. Tests were considered positive only if they were
positive before revascularization and reversed to nega-
tive after revascularization. Among the 21 patients with
an FFR < 0.75, only four showed concordant results of
noninvasive tests. Abbreviation: FFR, fractional flow
reserve. Source: From Ref. 23.
Figure 27.8 Example of two patients presenting with different clinical syndromes, and in whom an angiographically similar stenosis is found
in the proximal left anterior descending coronary artery. In the left example, the lesion has no hemodynamic significance and does not need
any form of mechanical revascularization. In the right example, the stenosis is hemodynamically very significant and deserves an intervention.
Abbreviation: FFR, fractional flow reserve.
of a nonsignificant stenosis in the left main may lead to atresia of In addition, angiographic assessments of left main lesions with
the conduits, even when internal mammary arteries are used an FFR < 0.75 were no different from those with an FFR > 0.75,
(33). Furthermore, the left main is among the most difficult further reinforcing the importance of physiological parameters
segments to assess by angiography (34). Noninvasive testing is in case of doubt. Therefore, patients with an intermediate left
often noncontributive in patients with a left main stenosis. main stenosis deserve physiological assessment before blindly
Perfusion defects are often seen in only one vascular territory, making a decision about the need for revascularization. Two
especially when the right coronary artery (RCA) is significantly examples shown in Figure 27.9 illustrate how FFR measurements
diseased (35). In addition, tracer uptake may be reduced in all in the left main may drastically influence the type of treatment in
vascular territories (“balanced ischemia”) giving rise to false- these patients. Left main disease is rarely isolated. When tight
negative studies (36). Several studies have shown that FFR could stenoses are present in the LAD or in the left circumflex artery
be used safely in left main stenosis and that the decision not to (LCx) the presence of these lesions will tend to increase the FFR
operate on left main stenosis with an FFR > 0.75 is safe (37–39). measured across the left main. The influence of an LAD/LCx
[2/7/010/20:47:53] [341–354]
Figure 27.9 (See color insert ) Example of two patients in whom FFR measurements in an “intermediate” ostial left main stenosis changed
the therapeutic strategy. The first (upper panel ) represents a 67-year-old man with massive mitral regurgitation who was scheduled for
minimally invasive (port access) mitral valvuloplasty. The coronary angiogram showed an intermediate ostial left main stenosis. The FFR of
the left main stenosis was 0.69. Accordingly, this patient underwent conventional bypass surgery and mitral valvuloplasty via a median
sternotomy. The second (lower panel ) represents an 89-year-old man with critical aortic stenosis, referred for aortic valve replacement and
bypass surgery because of the presence of an ostial left main stenosis. FFR of the left main stem was 0.83. Accordingly, only a percutaneous
aortic valve implantation was performed and the left main stenosis was left alone. Abbreviation: FFR, fractional flow reserve.
lesion on the FFR value of the left main will depend on the large discrepancy between the anatomic description and the
severity of this distal stenosis but, even more, on the vascular actual severity of each stenosis. For example, a patient may
territory supplied by this distal stenosis. For example, if the distal have three-vessel disease on the basis of the angiogram, but
stenosis is in the proximal LAD, its presence will notably impact actually have only two hemodynamically significant stenoses;
the stenosis in the left main. If the distal stenosis is located in a vice versa, a patient can be considered as having one vessel
small second marginal branch, its influence on the left main disease of the RCA but actually have a hemodynamically
stenosis will be minimal (Fig. 27.10). significant stenosis of the left main. Figure 27.11 shows a
typical example of a patient in whom the RCA and the
Multivessel Disease LCx are critically narrowed and in whom the mid-LAD
Patients with multivessel disease actually represent a very shows a mild stenosis. Myocardial perfusion imaging showed
heterogeneous population. Their anatomical features (number a reversible perfusion defect in the inferolateral segments and
of lesions, their location, and their respective degree of com- a normal flow distribution in the segments supplied by the
plexity) may vary tremendously and have major implications LAD. In contrast, FFR shows that all three vessels are signi-
for the revascularization strategy. Moreover, there is often a ficantly narrowed but to a different extent. This has a
Figure 27.10 Drawing of a left main coronary stenosis

associated with a proximal left anterior descending
stenosis (left ) or with a stenosis in the distal circumflex
coronary artery (right ). The influence of a second steno-
sis distal on the fractional flow reserve value of the left
main will depend on the severity of this distal stenosis
and on the vascular territory supplied by this distal
stenosis.
[2/7/010/20:47:53] [341–354]
Figure 27.11 (See color insert ) Example of two patients with multivessel disease. (A) A 46-year-old man with stable angina and
angiographic three-vessel disease but functional two-vessel disease (LAD, RCA). (B) A 69- year-old man with severe angina. MPI showed
a reversible defect in the inferolateral segments. From the angiogram it is obvious that the RCA and the LCx are significantly narrowed (no
pressure measurements are needed). However, the mid- LAD stenosis, considered “nonsignificant” on the angiogram, appears to be
hemodynamically significant. This LAD stenosis was undetected by MPI because the uptake of tracer is notably worse in the LCx territory than
in the LAD territory. Abbreviations: LAD, left anterior descending artery; RCA, right coronary artery; MPI, myocardial perfusion imaging; LCx,
left circumflex artery; FFR, fractional flow reserve.
[2/7/010/20:47:53] [341–354]
major implication as far as revascularization is concerned. segments (49). These data support the use of FFR to evaluate
Preliminary FFR-guided revascularization strategies in patients stenoses remote from a recent myocardial infarction.
with multivessel disease were very encouraging (40–42).
Tailoring the revascularization according to the functional sig- Diffuse Disease
nificance of the stenoses rather than on their mere angiographic Histopathology studies and, more recently, intravascular ultra-
appearance may decrease costs and avoid the need for surgical sound have shown that atherosclerosis is diffuse in nature and
revascularization. A recent randomized multicenter study that a discrete stenosis in an otherwise normal artery is actually
(FAME) in 1000 patients showed that routine measurement of rare. The concept of a focal lesion is a mainly angiographic
FFR during PCI with drug-eluting stent(s) in patients with description but does not reflect pathology. Until recently, it was
multivessel disease, when compared with current angiography- believed that when no focal narrowing of >50% was seen at the
guided strategy reduces the rate of the composite end point of angiogram, no abnormal resistance was present in the epicar-
death, myocardial infarction, re-PCI, and coronary artery dial artery. It was therefore assumed that distal pressure was
bypass surgery at one year by approximately 30% and reduces normal and thus that “diffuse mild disease without focal
mortality and myocardial infarction at one year by approxi- stenosis” could not cause myocardial ischemia. This paradigm
mately 35%. Moreover, the FFR-guided strategy reduces the has recently been shifted: the presence of diffuse disease is
number of stents used, decreases the amount of contrast often associated with a progressive decrease in coronary pres-
agent used, does not prolong the procedure and is cost saving sure (7) and flow (50), and this cannot be predicted from the
(43,44). angiogram. In contrast this decline in pressure correlates with
the total atherosclerotic burden (51). In approximately 10% of
After Myocardial Infarction patients this abnormal epicardial resistance may be responsible
After a myocardial infarction, previously viable tissue is par- for reversible myocardial ischemia. In these patients chest pain
tially replaced by scar tissue. Therefore, the total mass of is often considered noncoronary because no single focal steno-
functional myocardium supplied by a given stenosis in an sis is found, and the myocardial perfusion imaging is wrongly
infarct related artery will tend to decrease (45). By definition, considered false positive (“false false positive”) (52). Such dif-
hyperemic flow and thus hyperemic gradient will both fuse disease and its hemodynamic impact should always be
decrease as well. Assuming that the morphology of the stenosis kept in mind when performing functional measurements. In a
remains identical, FFR must therefore increase. This does not large multicenter registry of 750 patients FFR was obtained
mean that FFR underestimates lesion severity after myocardial after technically successful stenting. A post-PCI FFR value <0.9
infarction. It simply illustrates the relationship that exists was still present in almost one-third of patients and was
between flow, pressure gradient and myocardial mass and, associated with a poor clinical outcome (53). The only way to
conversely, illustrates that the mere morphology of a stenotic demonstrate the hemodynamic impact of diffuse disease is to
segment does not necessarily reflect its functional importance. perform a careful pullback maneuver of the pressure sensor
This principle is illustrated in Figure 27.12. Recent data confirm under steady-state maximal hyperemia (Fig. 27.13).
that the hyperemic myocardial resistance in viable myocardium
Sequential Stenoses
within the infarcted area remains normal (46). This further
When several stenoses are present in the same artery, the
supports the application of the established FFR cutoff value
concept and the clinical value of FFR is still valid to assess
in the setting of partially infarcted territories. Earlier data had
the effect of all stenoses together. Yet, it is important to realize
suggested that microvascular function was abnormal in regions
that when several discrete stenoses are present in the same
remote from a recent myocardial infarction (47,48). However,
coronary artery, each of them will influence hyperemic flow
more recent work taking into account distal coronary pressure
and therefore the pressure gradient across the other one. The
indicates that hyperemic resistance is normal in these remote
influence of the distal lesion on the proximal is more important
than the reverse. The FFR can theoretically be calculated for
each stenosis individually (54,55). However, this is neither
practical nor easy to perform and therefore of little use in the
catheterization laboratory. Practically, as for diffuse disease, a
pullback maneuver under maximal hyperemia is the only way
to appreciate the exact location and physiological significance
of sequential stenoses.
Bifurcation Lesions
Overlapping of vessel segments as well as radiographic arti-
facts render bifurcation stenoses particularly difficult to evalu-
ate at angiography, while PCI of bifurcations is often more
challenging than for regular stenoses. The principle of FFR-
guided PCI applies in bifurcation lesions even though clinical
outcome data are currently limited. Two recent studies by Koo
et al. (56,57) used FFR in the setting of bifurcation stenting. The
results of these studies can be summarized as follows: (i) After
Figure 27.12 Schematic representation of the relationship between stenting, the ostium of the side branch looks often “pinched.”
FFR and myocardial mass before and after myocardial infarction. Yet such stenoses are grossly overestimated by angiography—
Abbreviations: DS, diameter stenosis; FFR, fractional flow reserve. none of these ostial lesions where the diameter stenosis was
estimated as 75% were found to have an FFR below 0.75.
[2/7/010/20:47:53] [341–354]
Figure 27.13 (See color insert ) A 73-year-old man with angina related to a tight stenosis in the proximal right coronary artery. The distal
Pd/Pa ratio in the LAD artery is 0.74. The pressure pullback tracing under steady-state maximal hyperemia shows that the distal pressure
increases progressively in three or four “steps.” This indicates that the abnormal FFR value is due to diffuse disease, rather than to one focal
stenosis that was the intended target of percutaneous coronary intervention. Abbreviations: LAD, left anterior descending artery; FFR,
fractional flow reserve.
(ii) When kissing balloon dilation was performed only in ostial mammary arteries placed on mildly diseased native arteries
stenoses with an FFR < 0.75, the FFR at six months was >0.75 in showed a very high attrition rate (33).
95% of all cases.
FFR and Coronary Artery Bypass Graft Lesions CONCLUSION: TOWARD A NEW DIAGNOSTIC
Assessment of stenosis severity in coronary artery bypass grafts by ALGORITHM FOR PATIENTS WITH KNOWN OR
FFR is technically very easy. In addition, all theoretical assump- SUSPECTED CORONARY ARTERY DISEASE
tions underlying the concept of FFR hold in cases of bypasses. Pressure-derived FFR is a theoretically robust and practically
There is no reason to believe that another threshold value should simple means of assessing the functional consequences of
be found even though this has not being formally investigated in epicardial coronary atherosclerosis. With minimal experience
large series (58). Stated another way, FFR is able to define whether the technique of FFR measurements is simple, swift and safe.
or not a stenosis in a bypass graft is capable of inducing ischemia. Only a pressure wire and a bolus of hyperemia-inducing med-
In contrast, there are no clinical outcome data obtained in patients ication are required. Its invasive nature is counterbalanced by
in whom decisions regarding revascularization have been based its unequaled spatial resolution, offering functional information
on FFR. It seems common sense to admit that in patients with an down to the “per centimeter” level, while noninvasive tests
FFR < 0.75, the revascularization of this lesion might be beneficial operate, at best, at the “per vascular territory” level. Clinical
for the patient. However, deferring the revascularization proce- outcome data of patients in whom the revascularization strat-
dure on the basis of an FFR value >0.75 is not proven. Therefore, egy has been based on FFR measurements are very encourag-
FFR should be used with caution in bypass grafts. ing. Accordingly, FFR can be considered as the interventional
One important study by Botman et al. has investigated cardiologists’ “pocket myocardial perfusion imaging” modal-
the relationship between stenosis severity in native arteries and ity. This is true with some important qualifications: (i) FFR is
graft patency after six months. The authors showed that the more accurate in intermediate lesions; (ii) FFR has a better
rate of occlusion was approximately three times higher when spatial resolution; (iii) combined with the index of myocardial
the bypass was placed on a native artery with a hemodynami- resistance, FFR is able to distinguish epicardial and myocardial
cally nonsignificant stenosis (59). These results corroborate resistance (10); and (iv) it is available in the catheterization
the data reported by Berger et al. who showed that internal laboratory, as FFR is performed in conjunction with coronary
[2/7/010/20:47:53] [341–354]
angiography. Therefore, it is the only true “all-in-one” approach 10. Fearon WF, Balsam LB, Farouque HM, et al. Novel index for
for patients with suspected or known coronary artery disease as invasively assessing the coronary microcirculation. Circulation
it combines unequaled physiological information, the best possi- 2003; 107:3129–3132.
ble anatomical information, and the possibility of immediate 11. Fearon WF, Shah M, Ng M, et al. Predictive value of the index of
microcirculatory resistance in patients with ST-segment elevation
revascularization if needed.
myocardial infarction. J Am Coll Cardiol 2008; 51:560–565.
We believe that the diagnostic work-up of patients with 12. Sezer M, Oflaz H, Goren T, et al. Intracoronary streptokinase after
known or suspected coronary artery disease might be drasti- primary percutaneous coronary intervention. N Engl J Med 2007;
cally shortened. The conventional teaching (60) is that patients 356:1823–1834.
with suspected coronary stenosis should first undergo non- 13. Siebes M, Verhoeff BJ, Meuwissen M, et al. Single-wire pressure
invasive functional testing, the so-called “gatekeepers” of the and flow velocity measurement to quantify coronary stenosis
catheterization laboratory. Patients are referred for diagnostic hemodynamics and effects of percutaneous interventions. Circu-
coronary angiography if (and only if) these tests indicate lation 2004; 109:756–762.
reversible myocardial ischemia. At angiography, a 50% to 14. Meuwissen M, Siebes M, Chamuleau SA, et al. Hyperemic stenosis
70% stenosis is often considered a justification for revasculari- resistance index for evaluation of functional coronary lesion
severity. Circulation 2002; 106:441–446.
zation. We have seen, however, how often the results of non-
15. Pijls NH, van Son JA, Kirkeeide RL, et al. Experimental basis of
invasive functional tests performed sequentially are inaccurate determining maximum coronary, myocardial, and collateral blood
and/or contradictory. In addition, the angiographic degree of flow by pressure measurements for assessing functional stenosis
stenosis is a battered gold standard, leading to a large number severity before and after percutaneous transluminal coronary
of inappropriate decisions regarding revascularization (30). angioplasty. Circulation 1993; 87:1354–1367.
Finally, noninvasive testing is actually performed in only a 16. De Bruyne B, Baudhuin T, Melin JA, et al. Coronary flow reserve
minority of patients undergoing angioplasty (29). calculated from pressure measurements in humans. Validation
In contrast to this conventional approach, we propose to with positron emission tomography. Circulation 1994; 89:
put more emphasis on a careful interrogation of the patients 1013–1022.
including a precise analysis of his/her risk factors. If, on this 17. Legalery P, Seronde MF, Meneveau N, et al. Measuring pressure-
derived fractional flow reserve through four French diagnostic
basis, an experienced cardiologist comes to the conclusion that
catheters. Am J Cardiol 2003; 91:1075–1078.
“this person might well have significant coronary artery sten- 18. De Bruyne B, Pijls NH, Barbato E, et al. Intracoronary and intra-
oses,” we believe it is more efficacious to send the patient directly venous adenosine 50 -triphosphate, adenosine, papaverine, and
to the catheterization laboratory if and only if in the catheter- contrast medium to assess fractional flow reserve in humans.
ization laboratory, FFR measurements can be obtained, and the Circulation 2003; 107:1877–1883.
revascularization strategy is guided by the integration of clinical, 19. McGeoch RJ, Oldroyd KG. Pharmacological options for inducing
anatomic (angiographic), and functional (FFR) information. maximal hyperaemia during studies of coronary physiology.
Catheter Cardiovasc Interv 2008; 71:198–204.
20. Parham WA, Bouhasin A, Ciaramita JP, et al. Coronary hyperemic
REFERENCES dose responses of intracoronary sodium nitroprusside. Circulation
1. Verma S, Anderson TJ. Fundamentals of endothelial function for 2004; 109:1236–1243.
the clinical cardiologist. Circulation 2002; 105:546–549. 21. Pijls NH, Van Gelder B, Van der Voort P, et al. Fractional flow
2. Seiler C, Kirkeeide RL, Gould KL. Basic structure-function rela- reserve. A useful index to evaluate the influence of an epicardial
tions of the epicardial coronary vascular tree. Basis of quantitative coronary stenosis on myocardial blood flow. Circulation 1995; 92:
coronary arteriography for diffuse coronary artery disease. Circu- 3183–3193.
lation 1992; 85:1987–2003. 22. De Bruyne B, Bartunek J, Sys SU, et al. Relation between myo-
3. Finet G GM, Perrenot B, Rioufol G, et al. Fractal geometry of cardial fractional flow reserve calculated from coronary pressure
arterial coronary bifurcations: a quantitative coronary angiogra- measurements and exercise-induced myocardial ischemia. Circu-
phy and intravascular ultrsound analysis. EuroIntervention 2007; lation 1995; 92(1):39–46.
3:490–498. 23. Pijls NH, De Bruyne B, Peels K, et al. Measurement of fractional
4. Gazetopoulos N, Ioannidis PJ, Marselos A, et al. Length of main flow reserve to assess the functional severity of coronary-artery
left coronary artery in relation to atherosclerosis of its branches. A stenoses. N Engl J Med 1996; 334:1703–1708.
coronary arteriographic study. Br Heart J 1976; 38:180–185. 24. Bartunek J, Marwick TH, Rodrigues AC, et al. Dobutamine-
5. Costa RA, Mintz GS, Carlier SG, et al. Bifurcation coronary lesions induced wall motion abnormalities: correlations with myocardial
treated with the “crush” technique: an intravascular ultrasound fractional flow reserve and quantitative coronary angiography. J
analysis. J Am Coll Cardiol 2005; 46:599–605. Am Coll Cardiol 1996; 27(6):1429–1436.
6. Canty JM Jr. Coronary pressure-function and steady-state pres- 25. Chamuleau SA, Meuwissen M, van Eck-Smit BL, et al. Fractional
sure-flow relations during autoregulation in the unanesthetized flow reserve, absolute and relative coronary blood flow velocity
dog. Circ Res 1988; 63:821–836. reserve in relation to the results of technetium-99m sestamibi
7. De Bruyne B, Hersbach F, Pijls NH, et al. Abnormal epicardial single-photon emission computed tomography in patients with
coronary resistance in patients with diffuse atherosclerosis but two-vessel coronary artery disease. J Am Coll Cardiol 2001; 37(5):
“normal” coronary angiography. Circulation 2001; 104:2401–2406. 1316–1322.
8. Gould KL, Lipscomb K, Hamilton GW. Physiologic basis for 26. Abe M, Tomiyama H, Yoshida H, et al. Diastolic fractional flow
assessing critical coronary stenosis. Instantaneous flow response reserve to assess the functional severity of moderate coronary
and regional distribution during coronary hyperemia as measures artery stenoses: comparison with fractional flow reserve and cor-
of coronary flow reserve. Am J Cardiol 1974; 33:87–94. onary flow velocity reserve. Circulation 2000; 102(19):2365–2370.
9. Kern MJ, Lerman A, Bech JW, et al. Physiological assessment of 27. de Bruyne B, Bartunek J, Sys SU, et al. Simultaneous coronary
coronary artery disease in the cardiac catheterization laboratory: a pressure and flow velocity measurements in humans. Feasibility,
scientific statement from the American Heart Association Com- reproducibility, and hemodynamic dependence of coronary flow
mittee on Diagnostic and Interventional Cardiac Catheterization, velocity reserve, hyperemic flow versus pressure slope index, and
Council on Clinical Cardiology. Circulation 2006; 114:1321–1341. fractional flow reserve. Circulation 1996; 94:1842–1849.
[2/7/010/20:47:53] [341–354]
28. Murtagh B, Higano S, Lennon R, et al. Role of incremental doses of 44. Tonino PA, De Bruyne B, Pijls NH, et al. Fractional flow reserve
intracoronary adenosine for fractional flow reserve assessment. versus angiography for guiding percutaneous coronary interven-
Am Heart J 2003; 146:99–105. tion. N Engl J Med 2009; 360:213–224.
29. Topol EJ, Ellis SG, Cosgrove DM, et al. Analysis of coronary 45. De Bruyne B, Pijls NH, Bartunek J, et al. Fractional flow reserve in
angioplasty practice in the United States with an insurance-claims patients with prior myocardial infarction. Circulation 2001; 104:
data base. Circulation 1993; 87:1489–1497. 157–162.
30. Wijns W, De Bruyne B, Vanhoenacker PK. What does the clinical 46. Marques KM, Knaapen P, Boellaard R, et al. Microvascular func-
cardiologist need from noninvasive cardiac imaging: is it time to tion in viable myocardium after chronic infarction does not influ-
adjust practices to meet evolving demands? J Nucl Cardiol 2007; ence fractional flow reserve measurements. J Nucl Med 2007;
14:366–370. 48:1987–1992.
31. Pijls NH, van Schaardenburgh P, Manoharan G, et al. Percuta- 47. Uren NG, Crake T, Lefroy DC, et al. Reduced coronary vasodilator
neous coronary intervention of functionally nonsignificant steno- function in infarcted and normal myocardium after myocardial
sis: 5-year follow-up of the DEFER Study. J Am Coll Cardiol 2007; infarction. N Engl J Med 1994; 331:222–227.
49:2105–2111. 48. Claeys MJ, Vrints CJ, Bosmans J, et al. Coronary flow reserve
32. Chaitman BR, Fisher LD, Bourassa MG, et al. Effect of coronary during coronary angioplasty in patients with a recent myocardial
bypass surgery on survival patterns in subsets of patients with left infarction: relation to stenosis and myocardial viability. J Am Coll
main coronary artery disease. Report of the Collaborative Study in Cardiol 1996; 28:1712–1719.
Coronary Artery Surgery (CASS). Am J Cardiol 1981; 48:765–777. 49. Marques KM, Knaapen P, Boellaard R, et al. Hyperaemic micro-
33. Berger A, MacCarthy PA, Siebert U, et al. Long-term patency of vascular resistance is not increased in viable myocardium after
internal mammary artery bypass grafts: relationship with preop- chronic myocardial infarction. Eur Heart J 2007; 28:2320–2325.
erative severity of the native coronary artery stenosis. Circulation 50. Gould KL, Nakagawa Y, Nakagawa K, et al. Frequency and
2004; 110:II36–II40. clinical implications of fluid dynamically significant diffuse coro-
34. Lindstaedt M, Spiecker M, Perings C, et al. How good are expe- nary artery disease manifest as graded, longitudinal, base-to-apex
rienced interventional cardiologists at predicting the functional myocardial perfusion abnormalities by noninvasive positron
significance of intermediate or equivocal left main coronary artery emission tomography. Circulation 2000; 101:1931–1939.
stenoses? Int J Cardiol 2007; 120:254–261. 51. Fearon WF, Nakamura M, Lee DP, et al. Simultaneous assessment
35. Lima RS, Watson DD, Goode AR, et al. Incremental value of of fractional and coronary flow reserves in cardiac transplant
combined perfusion and function over perfusion alone by gated recipients: Physiologic Investigation for Transplant Arteriopathy
SPECT myocardial perfusion imaging for detection of severe (PITA Study). Circulation 2003; 108:1605–1610.
three-vessel coronary artery disease. J Am Coll Cardiol 2003; 42: 52. Aarnoudse WH, Botman KJ, Pijls NH. False-negative myocardial
64–70. scintigraphy in balanced three-vessel disease, revealed by coro-
36. Ragosta M, Bishop AH, Lipson LC, et al. Comparison between nary pressure measurement. Int J Cardiovasc Intervent 2003; 5:
angiography and fractional flow reserve versus single-photon 67–71.
emission computed tomographic myocardial perfusion imaging 53. Pijls NH, Klauss V, Siebert U, et al. Coronary pressure measure-
for determining lesion significance in patients with multivessel ment after stenting predicts adverse events at follow-up: a multi-
coronary disease. Am J Cardiol 2007; 99:896–902. center registry. Circulation 2002; 105:2950–2954.
37. Bech GJ, Droste H, Pijls NH, et al. Value of fractional flow reserve 54. De Bruyne B, Pijls NH, Heyndrickx GR, et al. Pressure-derived
in making decisions about bypass surgery for equivocal left main fractional flow reserve to assess serial epicardial stenoses:
coronary artery disease. Heart 2001; 86:547–552. theoretical basis and animal validation. Circulation 2000; 101:
38. Jasti V, Ivan E, Yalamanchili V, et al. Correlations between frac- 1840–1847.
tional flow reserve and intravascular ultrasound in patients with 55. Pijls NH, De Bruyne B, Bech GJ, et al. Coronary pressure mea-
an ambiguous left main coronary artery stenosis. Circulation 2004; surement to assess the hemodynamic significance of serial sten-
110:2831–2836. oses within one coronary artery: validation in humans. Circulation
39. Leesar MA, Mintz GS. Hemodynamic and intravascular ultra- 2000; 102:2371–2377.
sound assessment of an ambiguous left main coronary artery 56. Koo BK, Kang HJ, Youn TJ, et al. Physiologic assessment of jailed
stenosis. Catheter Cardiovasc Interv 2007; 70:721–730. side branch lesions using fractional flow reserve. J Am Coll
40. Botman KJ, Pijls NH, Bech JW, et al. Percutaneous coronary Cardiol 2005; 46:633–637.
intervention or bypass surgery in multivessel disease? A tailored 57. Koo BK, Park KW, Kang HJ, et al. Physiological evaluation of the
approach based on coronary pressure measurement. Catheter provisional side-branch intervention strategy for bifurcation
Cardiovasc Interv 2004; 63:184–191. lesions using fractional flow reserve. Eur Heart J 2008; 29:726–732.
41. Berger A, Botman KJ, MacCarthy PA, et al. Long-term clinical 58. Aqel R, Zoghbi GJ, Hage F, et al. Hemodynamic evaluation of
outcome after fractional flow reserve-guided percutaneous coro- coronary artery bypass graft lesions using fractional flow reserve.
nary intervention in patients with multivessel disease. J Am Coll Catheter Cardiovasc Interv 2008; 72:479–485.
Cardiol 2005; 46:438–442. 59. Botman CJ, Schonberger J, Koolen S, et al. Does stenosis severity of
42. Wongpraparut N, Yalamanchili V, Pasnoori V, et al. Thirty-month native vessels influence bypass graft patency? A prospective
outcome after fractional flow reserve-guided versus conventional fractional flow reserve-guided study. Ann Thorac Surg 2007;
multivessel percutaneous coronary intervention. Am J Cardiol 83:2093–2097.
2005; 96:877–884. 60. Bugiardini R, Bairey Merz CN. Angina with “normal” coronary
43. Fearon WF, Tonino PA, De Bruyne B, et al. Rationale and design of arteries: a changing philosophy. JAMA 2005; 293:477–484.
the Fractional Flow Reserve versus Angiography for Multivessel
Evaluation (FAME) study. Am Heart J 2007; 154:632–636.
[2/7/010/9:3:52] [355–369]
28
Intravascular ultrasound and virtual histology

Adriano Caixeta, Akiko Maehara, and Gary S. Mintz
INTRODUCTION Several clinically relevant properties of the ultrasound image—

Real-time ultrasound imaging originated in the late 1960s and such as the resolution, depth of penetration, and attenuation of
early 1970s when Bom et al. (1) pioneered the development of the acoustic—are dependent on the geometric and frequency
linear array transducers for visualizing cardiac chambers and properties of the transducer. The higher the center frequency,
valves. The first transluminal images of human arteries were the better the axial resolution, but the lower the depth of
recorded by Yock et al. (2) in 1988, when a miniaturized and penetration. For coronary imaging because the transducer is
single-transducer system was placed within the coronary close to the vessel wall, high ultrasound frequencies are used
arteries. Ever since, intravascular ultrasound (IVUS) has that are centered at 20 to 45 MHz (4). Axial resolution is
become an increasingly important catheter-based imaging tech- typically 80 to 120 mm, and lateral resolution is typically 100
nology providing practical guidance for percutaneous coronary to 250 mm (3).
intervention (PCI) and numerous clinical and research insights
(3,4). For example, IVUS directly images the coronary arterial Equipment
wall, which contains the major components of atherosclerotic Two different transducer designs are used: mechanically
plaque, allowing measurement of plaque size, distribution, and rotated and electronically-activated phased-array. Mechanical
to some extent its composition. Thus, IVUS has become estab- probes use a drive cable to rotate a single-element transducer at
lished as the method of choice for the serial assessment of the tip of the catheter at 1,800 rpm. At approximately
atherosclerotic plaque burden in progression-regression trials. 18 increments, the transducer sends and receives ultrasound
Recently, virtual histology IVUS (VH-IVUS) using spectral signals providing 256 individual radial scan lines for each
analysis of radiofrequency ultrasound backscatter has emerged image. In electronic systems, multiple tiny transducer elements
with the potential to better assess real-time plaque components in an annular array are activated sequentially to generate the
and vulnerable plaque (5). cross-sectional image (3,6).
This chapter will examine the rationale, technique, and
interpretation of IVUS and VH-IVUS imaging in diagnostic and
therapeutic applications. Imaging Artifacts
The recognition of imaging an artifact is critical because it may
interfere with image interpretation and measurements. The
GREYSCALE IVUS most common imaging artifacts are (i) ring-down, (ii) nonuni-
IVUS Imaging: Definitions and Basics form rotational distortion (NURD), and (iii) reverberations.
Ultrasound is acoustic energy with a frequency above human Ring-down artifacts are usually observed as a series of parallel
hearing. The highest frequency that the human ear can detect is bands or halos of variable thickness surrounding the catheter
approximately 20 thousand cycles per second (20,000 Hz). For and obscuring near-field imaging. Phased-array systems tend
medical diagnostic purposes, ultrasound imaging frequencies to have more ring-down artifacts. NURD is an artifact that
are in the range of millions of cycles per second (megahertz, arises from frictional forces to the rotating elements in mechan-
MHz). ical catheters. NURD creates stretched or compacted portions
The IVUS transducer converts electrical energy into of the images. Because accurate reconstruction of IVUS two-
acoustical energy through a piezoelectric (pressure-electric) dimensional images is dependent on uniform rotation of the
crystalline material that expands and contracts to produce catheter, NURD may create errors during IVUS measurements
sound waves when electrically excited. After reflection from (7). NURD artifacts can also occur because of bends in the
tissue, part of the ultrasound energy returns to the transducer; catheter driveshaft or in the presence of acute bends in the
the transducer then generates an electrical impulse that is artery. Reverberations are false repetitive echoes of the same
converted into moving pictures (6). All materials in the body structure that give the impression of second or various inter-
reflect sound waves. Sound waves bounce back at various faces at fixed-multiple distances from the transducer. Rever-
intervals depending on the type of material and the distance beration artifacts are more common from strong echo reflectors
from the transducer. It is the variation in reflective sound such as stents, guidewires, guiding catheters, and calcium
waves that creates the ultrasound image on the console. (especially after rotational atherectomy). There are a few
The intensity of reflected (or backscattered) ultrasound other artifacts that can also interfere in IVUS interpretation
depends on a number of variables including the intensity of the including side lobes and ghost artifacts also generated from
transmitted signal, the attenuation of the signal by the tissue, strong echo reflectors such as calcium and stent metal (3). In
the distance from the transducer to the target, the angle of the longitudinal or L-mode display, catheter motion artifacts dur-
signal relative to the target, and the density of the tissue (3) ing the pullback may results in a “saw tooth” appearance.
[2/7/010/9:3:52] [355–369]
Catheter position also plays an important role in image

quality. Off-axis position of the catheter may alter vessel geom-
etry in an elliptical fashion to mislead the operator to overes-
timate the lumen and vessel area. Axial (antegrade-retrograde)
movement of the IVUS probe during the cardiac cycle scram-
bles consecutive image slices that may have implications for
three-dimensional reconstruction and attempts to assess coro-
nary artery compliance (8).
Image Acquisition and Presentation

Two important consensus documents have been published on
image acquisition and reporting of IVUS data: (i) Standards for
the Acquisition, Measurement, and Reporting of IVUS Studies:
A Report of the American College of Cardiology Task Force on
Clinical Expert Consensus Documents (6) and (ii) the Study
Figure 28.1 Normal coronary artery morphology in a cross-
Group on Intracoronary Imaging of the Working Group of sectional view. In the magnified image in the left, only the outer
Coronary Circulation and the Subgroup on Intravascular Ultra- bright adventitial layer is representative of the monolayered
sound of the Working Group of Echocardiography of the appearance. In the magnified image in the right, the bright inner
European Society of Cardiology (9). layer (intima), middle echolucent zone (media), and outer bright
IVUS supplements angiography by providing a tomo- layer (adventitia) are representative of the three-layered appear-
graphic perspective of lumen geometry and vessel wall struc- ance of IVUS. Abbreviation: IVUS, intravascular ultrasound.
ture. A longitudinal view (L-mode or long view) can be also
displayed, but this should be done only when using motorized
transducer pullback. Longitudinal representation of IVUS
images is useful for length measurements, for interpolation of
shadowed deep arterial structures (i.e., external elastic mem-
brane (EEM) behind calcium or stent metal) (6).
There are advantages and disadvantages using manual or and elastic tissue; it is 300 to 500 mm thick. The outer border of
motorized pullback; however, motorized pullback is usually the adventitia is also indistinct due to echo reflectivity similar
preferable. Using motorized transducer pullback allows assess- to the surrounding periadventitial tissues (11). Therefore, the
ment of lesion length, volumetric measurements, consistent and normal coronary artery is either (i) “monolayered” in cases of
systematic IVUS image acquisition among different operators, intimal thickness <100 mm because of IVUS catheter resolution
and uniform and reproducible image acquisition for multi- is <100 mm or (ii) “three-layered” to include a bright echo from
center and serial studies. the intima, a dark zone from the media, and bright surrounding
In standard image acquisition after anticoagulation and echoes from the adventitia (Fig. 28.1).
intracoronary nitroglycerin administration, the IVUS catheter
should be placed distal to the segment of interest (at least 10
mm of distal reference), and a continuous pullback to the aorta Quantitative Analysis
should be recorded. The preferred pullback speed is 0.5 mm/sec. IVUS imaging is not identical to a histologic imaging. In non-
stented lesions, two strong acoustic interfaces are well visual-
ized by ultrasound: the leading edge of the intima and the outer
Normal Coronary Artery Morphology border of the media. Therefore, two cross-sectional area (CSA)
Three layers of the coronary artery are visible by IVUS imaging: measurements can be defined: the lumen CSA and the media/
intima, media, and adventitia. Normal intima thickness adventitia CSA (or EEM CSA). The atheroma or plaque&media
increases with age, from a single endothelial cell at birth to a (P&M) complex is calculated as EEM minus lumen; the media
mean of 60 mm at five years to 220 to 250 mm at 30 to 40 years of cannot be measured as a distinct structure. Thus, complete
age (10). The definition of abnormal intimal thickness by IVUS quantification of a nonstented lesion is possible by tracing the
is still controversial; in general, the threshold of “normal EEM and lumen areas of the proximal reference, lesion, and
intimal thickness” is <300 mm (0.3 mm). The innermost layer distal reference; calculating derived measures, including min-
of the intima is relatively echogenic compared with the lumen imum and maximum EEM and lumen diameters, P&M area
and media and displayed on the screen as a single bright and thickness, and plaque burden (P&M divided by EEM); and
concentric echo. The lower ultrasound reflectance of the measuring lesion length (distance between the proximal and
media is due to its homogeneous smooth muscle cells distri- distal reference) (Fig. 28.2).
bution and smaller amounts of collagen, elastic tissue, and In stented vessels, the stent forms a third measurable
proteoglycans. The thickness of media histologically averages structure (stent CSA). It appears as bright points along the
200 mm, but medial thinning occurs in the presence of athero- circumference of the vessel. Complete quantification of a
sclerosis (11). The intima/media border is poorly defined stented lesion is possible by tracing the EEM and lumen areas
because the intimal layer reflects ultrasound more strongly of the proximal and distal reference and the EEM, lumen, and
than the media. Conversely, the media/adventitia border, con- stent areas of the stented lesion; calculating derived measures
sistent with the location of the EEM, is accurately defined [minimum and maximum EEM, stent, and lumen diameters;
because a step-up in echo reflectivity occurs without blooming. peristent P&M area and thickness; and intrastent intimal hyper-
The outermost layer, the adventitia, is composed of collagen plasia (IH, area and %IH)]; and measuring stent length.
[2/7/010/9:3:52] [355–369]
INTRAVASCULAR ULTRASOUND AND VIRTUAL HISTOLOGY 357
Figure 28.2 IVUS measurements in a nonstented artery. The proximal (Panel A) and distal reference (Panel C) and minimum lumen area
(Panel B) of the lesion are shown. The IVUS study is shown in duplicate: 1 unlabeled and 1 highlighted with lines to illustrate quantitative
analysis. The dashed line highlights each external elastic membrane cross-sectional area (EEM CSA), and the solid line indicates each lumen
interface (lumen CSA). The minimal lumen cross-sectional area (lumen CSA) at the lesion site is 2.1 mm2. Between the EEM CSA and lumen
CSA, the atheroma or plaque and media (P&M) complex is calculated. Abbreviation: IVUS, intravascular ultrasound.
Plaque Composition adventitia, and (iii) calcific plaques, characterized by the pres-
Atherosclerotic plaques are heterogeneous and contain a mix- ence of acoustic shadowing along with the brightest echoes and
ture of plaque components with different impedance (density). reverberations (Fig. 28.3).
IVUS imaging can categorize lesions into subtypes according to Intimal hyperplasia due to in-stent restenosis often
echodensity—typically by using the collagen-rich “bright” appears to have low echogenecity depending, in part, on age
adventitia as a reference. Three basic types of lesions are and adjunct therapies (i.e., brachytherapy).
distinguished according to plaque echogenicity: (i) “soft” or The identification of thrombus is difficult by IVUS. It may
hypoechoic plaque does not reflect much ultrasound and appear as lobulated hypoechoic mass within the lumen, scin-
appears dark with less echointensity compared to the adven- tillating echoes, a distinct interface between the presumed
titia, (ii) “hard” or hyperechoic, composed primarily of fibrous thrombus and underlying plaque, and blood flow through the
tissue with echogenecity equal or greater intensity than the thrombus (3).
Figure 28.3 Panel A shows an example of a predominantly soft plaque—a thin fibrous cap (small arrows) and lipid core underlying it; the
plaque is less bright than the adventitia (a). In Panel B, fibrous or hyperechoic plaque is as bright as or brighter than the adventitia (a) without
shadowing. In this eccentric plaque, the thickness of the media behind the thickest part of the plaque (b) is an artifact caused by attenuation of
the beam as it passes through the hyperechoic plaque. In reality, the media becomes thinner with increasing atherosclerosis. Note that the
media behind the thinnest part of the plaque is also thinner—without artifacts. The Panel C shows superficial calcium—defined as calcium
(a) that is closer to the intima than it is to the adventitia. Calcium shadows the deeper arterial structures; in this case, the arc of calcification is
*1808 (dashed line).
[2/7/010/9:3:52] [355–369]
Discrepancies Between IVUS and Angiography second definition defines positive remodeling as a lesion EEM
Coronary angiography significantly underestimates the pres- greater than the proximal reference EEM, intermediate remodel-
ence, severity, and extent of atherosclerosis compared to IVUS ing as a lesion EEM between the proximal and distal reference
(12). Furthermore, IVUS routinely shows significant atheroscle- EEM, and negative remodeling as a lesion EEM less than the
rosis in angiographically “normal” segments in patients under- distal reference EEM. Using a third definition, arterial remodel-
going PCI (13). This discrepancy may be explained by three ing has been calculated by a remodeling index (lesion/reference
major factors: (i) atherosclerosis is often diffusely distributed EEM); positive remodeling is an index >1.05, intermediate
involving long segments of the vessel containing no truly remodeling is an index of 0.95 to 1.05, and negative remodeling
normal reference segment for comparison, (ii) complex athero- is an index <0.95 (19). It is important to note that all of these
sclerotic plaques are not appreciated by the two-dimensional remodeling definitions are based on a comparison of the refer-
“silhouette,” and (iii) most importantly, the presence of arterial ence EEM and lesion EEM. Accordingly, because both reference
wall remodeling (6). and lesion sites may have undergone quantitative changes in
EEM during the atherosclerotic process, the evidence of remod-
Arterial Remodeling eling derived from this index is relative and indirect (6).
Arterial wall remodeling at the site of coronary plaques was
originally described by Glagov et al. (14) from necropsy exami- Unstable Lesions
nations and later validated in vivo by IVUS imaging (15). In patients with acute coronary syndromes (ACS), culprit lesions
“Positive,” “outward,” or “expansive” remodeling is defined more frequently exhibit positive remodeling and a large plaque
as an increased in arterial dimensions; and “negative,” area; conversely, patients with a stable clinical presentation more
“inward,” or “constrictive” remodeling is defined as a smaller frequently show negative remodeling and a smaller plaque area
arterial dimension. Positive remodeling occurs as a compensa- (19). Echolucent plaques are also more common in unstable than
tory increase in local vessel size in response to increasing in stable patients. In addition, unstable lesions have less calcium
plaque burden, especially during early stages of atherosclerosis than stable lesions; and when present, calcific deposits in unsta-
(16). An absolute reduction in lumen dimensions typically does ble lesions are small, focal, and deep (21). Plaque ruptures can
not occur until the lesion occupies, on average, an estimated occur with varying clinical presentations although they are more
40% to 50% of the area within the EEM (40–50% plaque burden) often associated with ACS (22). Multiple ruptured plaques have
(14). Conversely, negative remodeling has been implicated in been reported in patients with ACS; their prevalence, however, is
the development of native significant stenosis in the absence of the subject of controversy (23,24).
plaque accumulation (Fig. 28.4) (17).
A number of definitions of remodeling have been pro- Intermediate Lesions
posed (15–20). One definition compares the lesion EEM CSA to Coronary angiography underestimates stenosis severity most
the average of the proximal þ distal reference EEM CSA; positive markedly in arteries with a 50% to 75% plaque burden and in
remodeling is an index >1.0 and negative remodeling <1.0. A patients with multivessel disease (25,26). An IVUS-measured
Figure 28.4 These illustrations show an eccentric, calcific, and small plaque accumulation (a) leading to negative remodeling. Panels A and
B refer to proximal and distal vessel references and their respective longitudinal views (white arrows). In Panel B, notice how the vessel cross-
sectional area (or EEM) is smaller than both the proximal and distal vessels. The longitudinal view depicts clearly the artery shrinkage at the
lesion site. Abbreviation: EEM, external elastic membrane.
[2/7/010/9:3:52] [355–369]
Figure 28.5 (Left) Mildly diseased ostium in the left main by angiography (white arrow). (Right) IVUS diagnostic study showing an eccentric
stenotic plaque with a minimum lumen area of 4.4 mm2. Notice the superficial and deep calcium with shadowing (arrows). Abbreviation: IVUS,
intravascular ultrasound.
minimum luminal area (MLA) of >4.0 mm2 in a major despite thorough examination using multiple radiographic
(>3.0 mm) proximal epicardial artery excluding the left main projections. The use of IVUS allows accurate characterization
coronary artery (LMCA) has been a consensus criterion of a of unusual morphology: filling defects, aneurysms, and spon-
nonsignificant coronary artery stenosis since it correlates well taneous dissections. While most filling defects are true thrombi,
with the findings of other methods including single-photon a small percentage is highly calcified plaque (Fig. 28.6) or even
emission computed tomography (27), Doppler FloWire studies calcified nodules, an unusual form of vulnerable plaque.
(28), and pressure wire measurements (29). The clinical impor- In an IVUS analysis of 77 angiographically diagnosed
tance of this criterion has been confirmed by a study of aneurysms, 27% were true aneurysms (Fig. 28.7), 4% were
300 patients showing that deferral of revascularization is safe pseudoaneurysms (Fig. 28.8), 16% were complex plaques, and
for patients with an MLA of >4.0 mm2 (30). 53% were normal arterial segments adjacent to stenoses (40).
By IVUS, a spontaneous dissection appears as a medial
Left Main Coronary Artery Disease dissection with an intramural hematoma occupying some or all
Several studies have showed that a very high percentage of of the dissected false lumen without identifiable intimal tears
patients with angiographically normal LMCA have disease by and without a communication between the true and false
IVUS (31–33). Others have shown that IVUS is helpful in lumens, typically in a nonatherosclerotic artery.
assessing ambiguous LMCA disease (Fig. 28.5) (34,35). The
main reasons for the discrepancy between angiography and Percutaneous Coronary Intervention
IVUS are (34) Stent Sizing
1. diffuse atherosclerotic plaque involvement may lead to a Preinterventional IVUS is performed to assess stenosis severity
lack of a normal reference segment, and plaque composition and distribution, measure reference
2. a short LMCA makes identification of a normal reference vessel size, and measure lesion length. As a result stent size can
segment difficult, be chosen more accurately than solely by angiography. There
3. the presence of arterial remodeling, are a number of paradigms that can be used. Stent size can be
4. the correlation between angiography and necropsy or selected by identifying the maximum reference lumen diameter
IVUS appears to be better in non-LMCA lesions possibly (proximal or distal to the lesion); it results in stent upsizing
because of unique geometric issues in the LMCA, and without an increase in complications. At the other extreme,
5. significant inter and intraobserver variability in the angio- stents can be sized to the “true vessel,” “media-to-media,” or
graphic assessment of LMCA disease (36) especially in midwall dimensions to reflect the amount of angiographically
ostium location (37). silent disease and, in most cases, the extent of positive remod-
eling, not just vessel size. Typically, this measurement will be
IVUS assessment of lumen dimensions has been shown to larger than lumen reference and, thus, should be used only by
correlate with fractional flow reserve (35), and both IVUS and experienced operators who understand its limitations.
fractional flow reserve predict clinical outcome in patients with IVUS measures lesion length more accurately than
LMCA disease (34,38). Jasti et al. reported that an IVUS MLD and angiography because IVUS eliminates foreshortening, vessel
MLA of <2.8 mm and <5.9 mm2, respectively, strongly predict tortuosity, or bendpoints.
the physiological significance of LMCA stenosis (35). In general,
an LMCA MLA <6.0 mm2 has been used as a criteria of signif- Stent Expansion and Malapposition
icant stenosis because it is the best predictor of FFR <0.75 and IVUS studies have shown that lumen enlargement after stent
correlates with MLA >4.0 mm2 in the left anterior descending implantation is a combination of vessel expansion and plaque
(LAD) and left circumflex (LCX) using Murray’s Law (39). redistribution/embolization, not plaque compression (41,42).
Plaque reduction in patients with ACS is attributed to plaque or
Unusual Lesion Morphology thrombus embolization (43). Intrusion or prolapse of plaque
During coronary angiography, it is common to encounter through the stent mesh into the lumen is more common in
unusual appearing lesions that elude accurate characterization ACS and in saphenous vein graft lesions. Importantly, after
[2/7/010/9:3:52] [355–369]
Figure 28.6 Diagnostic IVUS was performed to assess this angiographic filling defect at the proximal right coronary artery (white arrow in the
angiogram). The IVUS imaging run begins at the ostium (0 mm) of the right coronary artery to beyond the filling defect (6.0 mm). Note the
calcification (2.0 mm; white arrow in the IVUS) without lumen compromise. Abbreviation: IVUS, intravascular ultrasound. Source: From Ref. 3.
Figure 28.7 This patient presented with a true saccular aneurysm in the right coronary artery. IVUS imaging shows the aneurysm (Panel A)
and the proximal vessel (Panel B). Note that the intima (i), the media (ii), and the adventitia (iii) are intact, making this a true aneurysm.
Abbreviation: IVUS, intravascular ultrasound. Source: From Ref. 3.
stent implantation, there is a significant residual plaque burden drug-eluting stent (DES) implantation]. In the majority of pre-
behind the stent struts that almost always measures 50% to 75% DES studies, IVUS use optimized stent expansion; and the
at the center of the lesion. Thus, the stent CSA always looks initially larger MSA achieved was associated with a lower
smaller than the EEM even when the stent is fully expanded. restenosis rate (44–54). In the bare metal stent (BMS) era, 10
Apposition refers to the contact between the stent struts of 12 studies supported IVUS-guided PCI (Table 28.1).
to the arterial wall (6). Incomplete stent apposition is defined as At the introduction of DES, the importance of optimal
one or more struts clearly separated from vessel wall with stent deployment was initially underestimated. Suboptimal
evidence of blood speckles behind the strut (Fig. 28.9). There is stent expansion with both BMS and DES was a risk factor for
no conclusive evidence suggesting that isolated acute incom- restenosis and target vessel revascularization, but also for stent
plete stent apposition (in the absence of concomitant under- thrombosis (62–64). Recently, Roy et al. reported that IVUS
expansion) is associated with adverse clinical outcomes. guidance during DES implantation had the potential to reduce
both DES thrombosis and the need for repeat revascularization
IVUS-Guided Stent Implantation (65). In this study, 884 patients undergoing IVUS-guided
The two main uses of IVUS are to insure optimal stent expan- DES implantation were compared with 884 propensity-
sion (stent CSA) and full coverage of the lesion [especially with score matched patients undergoing DES implantation with
[2/7/010/9:3:52] [355–369]
Table 28.1 Angiography Versus IVUS-Guided PCI in BMS Era

Angio IVUS IVUS better
Study better better and cheaper
Choi et al. (45) X
CENIC (55) X
CRUISE (56) X
SIPS (46, 57) X X
AVID (58) X
Gaster et al. (47, 59) X X
RESIST (48, 60) X
TULIP (49) X
BEST (61) X
OPTICUS (50) X
PRESTO (51) X
DIPOL (54) X
Abbreviations: IVUS, intravascular ultrasound; PCI, percutaneous
coronary intervention; BMS, bare metal stent.
Recognition of Complications
IVUS has a higher sensitivity than angiography in identifying
complications that may occur during PCI. Angiography tends
to underestimate the presence and extent of dissection. Edge
Figure 28.8 This patient underwent a previous directional coronary
atherectomy of a lesion, in the left anterior descending artery, during dissections may be more common when the stent ends in a
which the artery was perforated. Follow-up catheterization showed reference segment that contains (i) both plaque and normal
both restenosis and a large aneurysm on angiography. The IVUS vessel wall or (ii) both calcific (or hard) and soft plaque
shows the body of the aneurism (Panel A) and the eccentric prox- elements. Dissection may not be visible by IVUS if the true
imal restenotic lesion (Panel B). Notice that the adventitia stops at lumen is severely stenotic and the ultrasound catheter presses
the point of transition from the vessel to the aneurism (a), indicating the flap against the arterial wall or if the dissection occurs
loss of vessel wall integrity and making this, in fact, a pseudoaneur- behind a calcified plaque that prevents accurate morphological
ysm. Abbreviation: IVUS, intravascular ultrasound. Source: From definition. In general, the treatment of coronary dissection with
Ref. 3. stent implantation depends on the combination of angiographic
assessment, flow assessment, and signs or symptoms of ischemia,
and residual IVUS MLA. Treatment of dissections should
be based on IVUS findings when they show evidence
angiographic guidance alone. At 30 days and at 12 months, a (i) reduced lumen dimensions below the threshold for an
lower rate of definite stent thrombosis using the ARC definition optimum result, (ii) impingement of the dissection flap on the
was seen in the IVUS-guided group (0.5% vs. 1.4%; P ¼ 0.046) IVUS catheter, (iii) mobility, and (iv) increased length. In gen-
and (0.7% vs. 2.0%; P ¼ 0.014), respectively. At one year, target eral, minor edge dissections should not be treated unless they
lesion revascularization was also lower in the IVUS-guided result in lumen compromise; the vast majority have healed
group (5.1% vs. 7.2%; P ¼ 0.07). A multicenter Korean registry when imaged at follow-up.
of 805 LMCA interventions showed that the 595 patients treated Intramural hematoma is a variant of dissection. Blood
with IVUS-guided DES implantation had better three-year accumulates in the medial space; the EEM expands outward
survival than the 210 patients in whom IVUS was not used to and the internal elastic membrane is pushed inward to cause
guide LMCA DES implantation (HR ¼ 0.43, P ¼ 0.019). lumen compromise. Intramural hematomas are typically
Figure 28.9 The figure illustrates an example of acute stent malapposition. Panel A shows the vessel before stenting implantation; the IVUS
study in Panels B and C shows the stent malapposition. In the magnified image in the right, notice the space between the stent strut and the
intima and the blood speckle behind the stent struts (a). Five stent struts are malposed (white arrows). Because of stent malapposition, the
stent area (9.4 mm2) is smaller than the lumen area (14.4 mm2). Abbreviation: IVUS, intravascular ultrasound.
[2/7/010/9:3:52] [355–369]
Figure 28.10 This patient presented with in-stent restenosis. After balloon dilation, coronary perforation with myocardial contrast
extravasation occurred, as seen by angiography (arrows). On IVUS, notice the small vessel with stenting at a distal site (Panel D) and
the medial and adventitial discontinuation at the site of perforation (Panels B, C and arrows). Notice also an accumulation of blood outside the
EEM (a). One stent graft was implanted prior to IVUS assessment (Panel A), followed by an additional stent graft after IVUS assessment.
Abbreviations: IVUS, intravascular ultrasound; EEM, external elastic membrane.
hyperechoic and crescent shaped, with straightening of the fully absorbable DES showed a significant reduction in stent
internal elastic membrane (66). In general, an intramural hem- CSA at follow-up (11.8%), absence of vessel negative remod-
atoma should be treated because of the propensity for propa- eling, and IH of 30%. Of concern was the higher incidence of
gation and lumen compromise. late-acquired incomplete stent apposition of 27% (73).
Coronary perforation and rupture usually occurs with over- Stent underexpansion is a common finding in restenotic
aggressive and/or oversized balloon dilation although it can occur stents. It is the result of poor expansion at implantation, not
with a guidewire and stenting as well. In general, there are three chronic stent recoil. In an analysis of over 1000 patients with
distinct IVUS morphologic patterns indicating arterial rupture: BMS restenosis, 15% had an MSA <4.5 mm2 and 25% had an
(i) free blood speckle outside the EEM (Fig. 28.10), (ii) extramural MSA between 4.5 and 6.0 mm2. In addition, in 4.5% there were
hematoma—an accumulation of blood outside the EEM, and technical and mechanical complications of stent implantation
(iii) less common, a new periadventitial echolucent interface that contributed to the restenosis. Examples of mechanical
representing contrast extravasation (3). Acute management ranges complications have included (i) missing the lesion (e.g., an
from a conservative strategy of monitoring to prolonged balloon aorto-ostial stenosis), (ii) stent “crush,” (iii) having the stent
inflations, covered stents, and surgery. stripped off the balloon during the implantation procedure, or
(iv) DES fracture (Fig. 28.11) (74).
Serial IVUS Studies Acquired Late Stent Malapposition

Restenosis Late stent malapposition (LSM) is usually caused by regional
Serial IVUS studies have shown that the main mechanism of vessel positive remodeling (Fig. 28.12). LSM has been reported
restenosis in nonstented arteries is negative arterial remodeling in 4% to 5% after BMS implantation (76). Studies have sug-
(decrease in EEM area), not intimal hyperplasia (17). Conversely, gested a higher incidence of LSM after DES (especially after
in-stent restenosis is primarily due to neointimal proliferation, SES) (77–79). Hoffmann et al. studied the impact of LSM after
not chronic stent recoil. By IVUS, %IH (IH volume divided by SES implantation on four-year clinical events. This pooled IVUS
stent volume) has been shown to be consistent for each stent type. analysis from three randomized trials comparing SES with BMS
DES reduce restenosis by reducing IH from an average of 30% in showed that LSM at follow-up was more common after SES
BMS (67) to 3% to 5% in sirolimus-eluting stents (SESs) (68,69), to than after BMS (25% vs. 8.3%; P ¼ 0.001); however, major
6% in everolimus-eluting stents (70), to 8% to 13% in polymer- adverse cardiac event-free survival at four years was identical
based paclitaxel-eluting stents (70,71), and to 16% in zotarolimus- for those with and without LSM (11.1% vs. 16.3%, P ¼ 0.48),
eluting stents (72). and LSM was not a predictor for target lesion revascularization,
Serial IVUS analysis of the first-in-human ABSORB trial target vessel failure, or late stent thrombosis during the four-
of the bioabsorbable vascular solutions (BVS, Abbott Vascular) year follow-up period in either patient group (80).
[2/7/010/9:3:52] [355–369]
Figure 28.11 This patient presented with restenosis at follow-up after CypherTM sirolimus-eluting stent implantation in the right coronary
artery (arrows on angiogram). Note the stent fracture with acquired transection on fluoroscopy. On IVUS, all stent struts have been seen at
proximal (Panel A) and distal (Panel C) references, whereas at the fracture site (Panel B; minimal lumen area of 4.8 mm2), only one stent strut
has been seen (arrow). Abbreviation: IVUS, intravascular ultrasound.
Figure 28.12 This patient underwent Cypher sirolimus-eluting stent implantation in a right coronary stenosis. The final angiogram is shown
in Panel A. At follow-up (Panel B), there was a proximal and focal angiographic aneurysm (white arrows). Final (post-stent implantation) IVUS
image is shown in Panel C and the follow-up IVUS image is shown in Panel D. Note the late stent malapposition (a and b). At the site of
maximum stent malapposition (b), there has been an increase in EEM CSA from 17.8 to 28.9 mm2. The stent CSA (8.8 mm2) and the peristent
P&M (8.9 mm2) have not changed. Abbreviation: CSA, cross-sectional area. Source: From Ref. 75.
[2/7/010/9:3:52] [355–369]
Figure 28.13 The radiofrequency signal received by IVUS. Plaque

characterization can be explored by analyzing the frequency rather
than the amplitude of the IVUS radiofrequency data. Note that
similar amplitude can have distinct frequencies. Abbreviation: IVUS,
intravascular ultrasound.
Conversely, others have suggested that LSM can contrib- similarities in the amplitude (Fig. 28.13) (85). Accordingly,
ute to late stent thrombosis (81,82). Cook et al. (82) studied plaque characterization can be explored by analyzing the fre-
13 patients presenting with very late stent thrombosis (>1 year) quency rather than the amplitude of the IVUS radiofrequency
after DES implantation and compared them to 144 control data. VH-IVUS uses the raw RF signal and elaborate automatic
patients who did not experience stent thrombosis. Compared calibration technology (known as “blind deconvolution”),
with DES controls, patients with very late stent thrombosis had which normalizes for catheter-to-catheter and system-to-system
longer lesions and stents, more stents per lesion, and more stent variability, allowing standardized image interpretation and
overlap. Vessel CSA was significantly larger for the in-stent tissue characterization (86,87). The RF that underlies the con-
segment (28.6 11.9 mm2 vs. 20.1 6.7 mm2; P ¼ 0.03) in very ventional gray scale IVUS image is analyzed using a spectral
late stent thrombosis patients compared with DES controls, analysis and compared with a histological database (VH-IVUS,
denoting evidence of positive arterial remodeling. Although Volcano Corporation, Rancho Cordova, California, U.S.). Eight
IVUS was not performed at stent implantation in any patients spectral parameters (maximum power, frequency at maximum
of either group, incomplete stent apposition was more frequent power, minimum power, frequency at minimum power, slope,
(77% vs. 12%, P < 0.001) and maximal incomplete stent appo- intercept, mid-band fit, and integrated backscatter) are com-
sition area was larger (8.3 7.5 mm2 vs. 4.0 3.8 mm2; P ¼ puted and combined in a statistical model to assess plaque
0.03) in patients with very late stent thrombosis compared with composition in terms of four major plaque components: (i)
controls. fibrous, (ii) fibrofatty, (iii) necrotic core, and (iv) dense calcified
On the basis of these studies with conflicting data, it is tissue. Fibrous tissue is identified as a dark green group of
still speculative as to how to treat patients with IVUS findings pixels on the VH-IVUS screen; it is characterized by areas of
of LSM. However, it appears that there are two populations of densely packed collagen with little or no lipid accumulation
LSM: (i) routinely detected LSM that is modest in size and does and no evidence of macrophages indicating inflammatory
not appear to cause late events and (ii) large areas of LSM response. Fibrofatty tissue is identified as a light green group
(essentially aneurysms) that are seen in patients with late stent of pixels on the VH-IVUS screen; it is characterized by areas of
thrombosis. significant lipid accumulation (mainly extracellular) with no
necrotic tissue. Necrotic core (NC) tissue is identified as a red
group of pixels on the VH-IVUS screen; it is characterized by a
VIRTUAL HISTOLOGY IVUS high level of lipid with many necrotic cells and remnants of
Gray scale IVUS is the gold standard for in vivo evaluation of
dead lymphocytes, foam cells, and microcalcifications. There
atherosclerotic plaque, as well as lumen and vessel dimensions.
are few or no collagen fibers, the cellular matrix is not well
However, several studies have suggested that IVUS has limited
organized, and there is poor mechanical stability. Dense cal-
value for the identification of specific plaque components
cium is identified as a white pixel group on the VH-IVUS
(5,83,84).
screen; this tissue is characterized by areas of dense calcium
without adjacent necrosis (Fig. 28.14) (5,83,88). The VH-IVUS
Basic Principle and Imaging Interpretation approach has led to a significant increase in the sensitivity and
Virtual histology (VH) is a new imaging modality based on the specificity of IVUS for plaque characterization, particularly
IVUS technique for quantifying plaque composition. Gray scale lipid deposits. The sensitivity of gray scale IVUS for detecting
IVUS uses the amplitude of the reflected ultrasound signal to lipid deposits is reported to be as low as 46%. VH-IVUS, on the
generate an image. Different tissues might have similar RF other hand, has a combined predictive accuracy >93% for all four
amplitude data, leading to misinterpretation of the gray scale VH plaque components, and very high sensitivity (72–96%), and
image. Nevertheless, the power and frequency of the RF signal specificity (91–99%), indicating good agreement with histology
commonly differs between tissues, regardless of eventual (88,89).
[2/7/010/9:3:52] [355–369]
Figure 28.14 (See color insert) Plaque composition and atherosclerotic plaque classification by VH-IVUS. (A) fibrotic, (B) fibrocalcific, (C)
pathological intimal thickening (D) thick cap fibroatheromas, and (E) thin cap fibroatheromas.
Data Acquisition TCFA is defined by the following: (i) confluent NC 10% of

VH-IVUS images are obtained with a commercially available plaque area in at least three consecutive frames without
console [IVG (InVision Gold) that has integral pcVH software overlying fibrous tissue and (ii) presence of 40% plaque
or S5 that has integral VIAS software, both Volcano Corpora- burden (plaque&media divided by EEM) (83). ThCFA is a
tion, Rancho Cordova, California, U.S.]. Both consoles are fibroatheroma with overlying fibrous tissue.
interfaced with either a phased-array transducer (Eagle EyeTM Correlation of in vivo VH-IVUS with histopathology also
20 MHz catheter, Volcano Corporation, Rancho Cordova, Cal- shows a high accuracy. Nasu et al. (88) analyzed 15 stable
ifornia, U.S.) or, in the future, with a developed mechanical angina pectoris patients and 15 ACS patients undergoing cor-
transducer (RevolutionTM 45 MHz, Volcano Corporation, Ran- onary atherectomy. The results of VH-IVUS data analysis
cho Cordova, California, U.S.). The integral software allows VH correlated well with histopathologic examination (predictive
analysis to be performed online in the catheterization labora- accuracy: 87.1% for fibrous, 87.1% for fibrofatty, 88.3% for NC,
tory. The IVUS catheter is positioned distally in the coronary and 96.5% for dense calcium regions). In addition, the authors
artery and then withdrawn at a speed of 0.5 mm/sec using an found that the frequency of NC was significantly higher in the
automated pullback device. VH-IVUS data are acquired simul- ACS group than in the stable group (in vitro histopathology:
taneously with the gray scale IVUS data; however, ECG-gating 22.6% vs. 12.6%, P ¼ 0.02; in vivo VH-IVUS: 24.5% vs. 10.4%,
is used to select frames for VH-IVUS analysis. P ¼ 0.002). Likewise, two different studies reported that vul-
nerable plaque by VH-IVUS occurred more often in patients
Atherosclerotic Lesion Phenotype Classification with ACS than in patients with stable angina. In one study,
On the basis on histopathological studies (90), coronary Rodriguez-Granilllo et al. (83) identified TCFA as a more
lesions are classified as (i) fibrotic, (ii) fibrocalcific, (iii) patho- prevalent finding in nonculprit, nonobstructive lesions in ACS
logical intimal thickening (PIT) (iv) thick cap fibroatheromas patients than in stable angina patients, with a larger percentage of
(ThCFA), and (v) thin cap fibroatheromas (TCFA) (Fig. 28.14). NC (12.26% vs. 7.40%, P ¼ 0.006) and less fibrotic tissue (63.96%
Fibrotic plaque is composed mainly of fibrous plaque with vs. 70.97%, P ¼ 0.007) (83). In another study, Hong et al. (95)
<10% of confluent NC, <15% of fibrofatty plaque, and <10% found that culprit lesions in patients with ACS had greater
of confluent dense calcium. Fibrocalcific plaque contains amounts of NC and smaller amounts of fibrofatty plaque com-
mainly fibrous plaques with >10% of confluent dense calcium. pared with culprit lesions in patients with stable angina.
PIT is a mixture of fibrous, fibrofatty, some necrotic core, and Potentially, the percentage of NC identified by VH-IVUS
some calcified tissue; PIT must have >15% of fibrofatty. Any can be used to stratify ACS patients. Missel et al. (96) showed that
plaque that contains >10% of confluent NC is classified as the percentage of NC and its ratio to dense calcium were posi-
fibroatheroma (ThCFA or TCFA). Pathology and autopsy tively associated with a high-risk ACS presentation. In this
studies indicate that TCFA is the most common type of vul- analysis, VH-IVUS was performed in culprit arteries of 225
nerable plaque (VP) and is a precursor of plaque rupture (91). patients with ACS; the percentage of NC and its ratio to dense
The vulnerability of plaque to rupture is typically character- calcium were associated with creatine kinase-MB levels, and was
ized by the presence of an NC, which is a region of the significantly higher for patients with ST-segment depression
fibroatheroma that is largely devoid of viable cells and con- compared with those with non-ST-segment depression ACS.
sists of cellular debris and cholesterol clefts, a thin fibrous cap Ruptured plaque debris and fragments of thrombotic
(<65 mm), and macrophage infiltration (92). The threshold for material appear to be responsible for distal embolization
defining the cap as “thin” (and categorize a lesion as a TCFA) during PCI in ACS, which is in turn associated with adverse
has been previously set at <65 mm, based on postmortem clinical outcomes. Some studies demonstrated that NC volume
studies. However, tissue shrinkage occurs during tissue fixa- predicted distal embolization in patients undergoing PCI for
tion, and postmortem contraction of arteries is an additional ST-segment elevation myocardial infarction (97) as well as in
confounding factor (93). Thus, the threshold used to define stable patients (98). Uetani et al. (98) showed that post-proce-
thin cap in vivo should probably be higher than 65 mm (83). dural myocardial necrosis was directly related to relative lipid
Indeed, some ex vivo studies have used higher (>200 mm) volume and inversely related to relative fibrous volume.
thresholds (94). Because the axial resolution of VH-IVUS is
from 100 to 150 mm, some experts maintain that the absence of Arterial Remodeling and Plaque Composition
visible fibrous tissue overlying a necrotic core suggests a cap There are conflicting data regarding the relationship of vessel
thickness of below 100 to 150 mm, and use the absence of such remodeling to plaque composition by VH-IVUS. In one study,
tissue to define a thin fibrous cap (83). Hence, by VH-IVUS, Rodriguez-Granillo et al. (99) showed that plaque composition
[2/7/010/9:3:52] [355–369]
Figure 28.15 (See color insert) Examples of conventional IVUS image and VH-IVUS side by side. (Left) Lesion with acute coronary
syndrome with positive remodeling. In this segment, notice the presence of thin cap fibroatheromas (i.e., necrotic core representing >10% of
the plaque burden without overlying fibrous tissue). (Right) Lesion with stable angina with negative remodeling. In this segment the plaque is
predominantly fibrotic. Abbreviations: IVUS, intravascular ultrasound; VH-IVUS, virtual histology IVUS.
and morphology by VH-IVUS analysis was associated with TCFA cannot be directly identified. Thus, vulnerable plaque
coronary remodeling. Positive remodeling was found with identification is compromised and might be imprecise by
high-risk lesions such as TCFA or fibroatheromas, whereas VH-IVUS.
negative remodeling was associated with less vulnerable
lesions, with 64% characterized by PIT, 29% as fibrocalcific
lesions, and only 7% fibroatheromatous lesions (P < 0.0001)
Potential Clinical Application
VH-IVUS could potentially aid in the assessment of changes in
(99). Surmely et al. (100) and Fujii et al. (101) reported the
plaque characteristics induced by lipid-lowering drugs (104),
opposite finding. NC was significantly smaller in lesions with
stenting, and could also help predict future coronary events.
positive remodeling than in lesions with intermediate/negative
VH-IVUS is a promising technique to detect vulnerable
remodeling. In one study (101), the authors showed a linear
plaque and to assess its natural history. The lesions that
relationship between remodeling index and fibrofatty plaque
harbor vulnerable plaque are frequently only mildly stenotic
area, demonstrating that positive remodeling occurs in lesions
on angiographic examination. Because VH-IVUS allows for
with more fibrofatty plaque. The discrepancy between these
improved interpretation of coronary disease in the entire
studies—including between the two latter studies with patho-
artery with volumetric data of the individual VH plaque
logic findings (102)—might have resulted from differences in
components, the method shows promise as an approach to
the patient/lesion population, in definition/methods applied
identify patients with a high-risk of adverse events. However,
or potential VH-IVUS artifacts/limitations. The Figure 28.15
there is currently no evidence favoring invasive treatment of a
shows two examples of plaque characterization by VH-IVUS.
vulnerable plaque. A meta-analysis showed that the restenosis
risk after balloon angioplasty or BMS implantation in angio-
VH-IVUS Limitations graphically determined intermediate lesions corresponded
The main limitations of VH-IVUS are as follows: (i) only four with the restenosis risk in clinically relevant stenosis and
plaque components are identifiable. For instance, there are no was unacceptably high (105). On the other hand, the risk of
classifications for thrombus or blood on VH-IVUS. Ruptured restenosis in intermediate/nonstenotic lesions treated with
plaques filled with blood or thrombus, or even luminal throm- DES is reported to be lower when compared with the esti-
bus, are assigned to 1 of the current plaque classifications, mated one-year probability of a nonfatal MI in lesions with a
usually fibrous. Thus, the lack of ability to identify thrombus stenosis <50% (106). Ongoing large-scale, prospective clinical
limits the recognition of certain high-risk plaques; (ii) ex vivo trials will help our understanding of the temporal changes
VH-IVUS validation studies have reported a high accuracy for that take place in coronary plaques and how VH-IVUS can be
the classification of plaque components based on selected used as a decision-making tool in individual patients. The
regions of interest representing homogeneous plaque compo- Providing Regional Observations to Study Predictors of
nents on the histology specimen. Validation studies for VH Events in the Coronary Tree (PROSPECT) trial is a natural
tissue maps of entire plaque cross-sections is reported to be history study designed to analyze the relationship of unex-
difficult due to the amorphous overlap of the tissue component pected ACS with the progression of coronary artery disease.
(89); therefore, VH-IVUS may have a limited accuracy to assess This first prospective trial using serial VH-IVUS analysis may
plaque components in highly heterogeneous and complex help to identify plaques that are prone to rupture, potentially
lesions (103); and finally (iii) there is no consensus about the leading to cardiac events. The results of this study could lead
in vivo definition of TCFA. The VH-IVUS definition is based on to changes in recommendations regarding how to treat vul-
the “absence” of visible fibrous tissue overlying a necrotic core; nerable plaque and coronary artery disease.
[2/7/010/9:3:52] [355–369]
CONCLUSIONS 14. Glagov S, Weisenberg E, Zarins CK, et al. Compensatory enlarge-

Gray scale IVUS provides (i) high-quality, tomographic imaging ment of human atherosclerotic coronary arteries. N Engl J Med
of the lumen, the atheroma, and the vessel wall, (ii) incremental 1986; 316:1371–1375.
and more detailed qualitative and quantitative information than 15. Hermiller JB, Tenaglia AN, Kisslo KB, et al. In vivo validation of
compensatory enlargement of atherosclerotic coronary arteries.
coronary angiography, (iii) practical guidance during PCI, and
Am J Cardiol 1993; 71:665–668.
(iv) numerous clinical and research insights. More recently, IVUS 16. Losordo DW, Rosenfield K, Kaufman J, et al. Focal compensatory
has become an indispensable part of DES research, as the most enlargement of human arteries in response to progressive athe-
effective tool in helping to understand the mechanisms, effects, rosclerosis. In vivo documentation using intravascular ultra-
and complications of this new stent technology. sound. Circulation 1994; 89:2570–2577.
VH-IVUS provides a reliable representation of plaque 17. Mintz GS, Kent KM, Pichard AD, et al. Contribution of inade-
composition and morphometry. Thin-capped fibroatheromas quate arterial remodeling to the development of focal coronary
can be identified using VH-IVUS. These lesions most often artery stenoses. An intravascular ultrasound study. Circulation
occur in ACS patients, and are associated with positive arterial 1997; 95:1791–1798.
remodeling. Necrotic core volume may also help identify lesions 18. Nishioka T, Luo H, Eigler NL, et al. Contribution of inadequate
compensatory enlargement to development of human coronary
most at risk for distal embolization during PCI. Prospective
artery stenosis: an in vivo intravascular ultrasound study. J Am
clinical trials will provide a better understanding of the temporal Coll Cardiol 1996; 27:1571–1576.
changes that occur in coronary plaques and how VH-IVUS can be 19. Schoenhagen P, Ziada KM, Kapadia SR, et al. Extent and direc-
used as a decision-making tool in individual patients. tion of arterial remodeling in stable versus unstable coronary
syndromes : an intravascular ultrasound study. Circulation 2000;
101:598–603.
REFERENCES 20. Pasterkamp G, Schoneveld AH, van der Wal AC, et al. Relation
1. Bom N, Lancee CT, Honkoop J, et al. Ultrasonic viewer for cross- of arterial geometry to luminal narrowing and histologic markers
sectional analyses of moving cardiac structures. Biomed Eng for plaque vulnerability: the remodeling paradox. J Am Coll
1971; 6:500-3–505. Cardiol 1998; 32:655–662.
2. Yock PG, Johnson EL, Linker DT. Intravascular ultrasound: 21. Mintz GS, Pichard AD, Popma JJ, et al. Determinants and
development and clinical potential. Am J Card Imaging 1988; correlates of target lesion calcium in coronary artery disease: a
2:185–193. clinical, angiographic and intravascular ultrasound study. J Am
3. Mintz G. Intracoronary Ultrasound. Oxon, UK: Taylor & Francis, Coll Cardiol 1997; 29:268–274.
2005. 22. Ge J, Chirillo F, Schwedtmann J, et al. Screening of ruptured
4. Honda Y, Fitzgerald PJ. Frontiers in Intravascular Imaging plaques in patients with coronary artery disease by intravascular
Technologies. Circulation 2008; 117:2024–2037. ultrasound. Heart 1999; 81:621–627.
5. Nair A, Kuban BD, Tuzcu EM, et al. Coronary plaque classifica- 23. Rioufol G, Finet G, Ginon I, et al. Multiple atherosclerotic plaque
tion with intravascular ultrasound radiofrequency data analysis. rupture in acute coronary syndrome: a three-vessel intravascular
Circulation 2002; 106:2200–2206. ultrasound study. Circulation 2002; 106:804–808.
6. Mintz GS, Nissen SE, Anderson WD, et al. American College of 24. Hong MK, Mintz GS, Lee CW, et al. Comparison of coronary
Cardiology Clinical Expert Consensus Document on Standards plaque rupture between stable angina and acute myocardial
for Acquisition, Measurement and Reporting of Intravascular infarction: a three-vessel intravascular ultrasound study in 235
Ultrasound Studies (IVUS). A report of the American College of patients. Circulation 2004; 110:928–933.
Cardiology Task Force on Clinical Expert Consensus Documents. 25. Arnett EN, Isner JM, Redwood DR, et al. Coronary artery
J Am Coll Cardiol 2001; 37:1478–1492. narrowing in coronary heart disease: comparison of cineangio-
7. Kimura BJ, Bhargava V, Palinski W, et al. Distortion of intra- graphic and necropsy findings. Ann Intern Med 1979; 91:350–356.
vascular ultrasound images because of nonuniform angular veloc- 26. Fernandes MR, Silva GV, Caixeta A, et al. Assessing intermediate
ity of mechanical-type transducers. Am Heart J 1996; 132:328–336. coronary lesions: angiographic prediction of lesion severity on
8. Arbab-Zadeh A, DeMaria AN, Penny WF, et al. Axial movement intravascular ultrasound. J Invasive Cardiol 2007; 19:412–416.
of the intravascular ultrasound probe during the cardiac cycle: 27. Nishioka T, Amanullah AM, Luo H, et al. Clinical validation of
implications for three-dimensional reconstruction and measure- intravascular ultrasound imaging for assessment of coronary
ments of coronary dimensions. Am Heart J 1999; 138:865–872. stenosis severity: comparison with stress myocardial perfusion
9. Di Mario C, Gorge G, Peters R, et al. Clinical application and imaging. J Am Coll Cardiol 1999; 33:1870–1878.
image interpretation in intracoronary ultrasound. Study Group 28. Abizaid A, Mintz GS, Pichard AD, et al. Clinical, intravascular
on Intracoronary Imaging of the Working Group of Coronary ultrasound, and quantitative angiographic determinants of the
Circulation and of the Subgroup on Intravascular Ultrasound of coronary flow reserve before and after percutaneous translumi-
the Working Group of Echocardiography of the European Soci- nal coronary angioplasty. Am J Cardiol 1998; 82:423–428.
ety of Cardiology. Eur Heart J 1998; 19:207–229. 29. Takagi A, Tsurumi Y, Ishii Y, et al. Clinical potential of intra-
10. Velican D, Velican C. Comparative study on age-related changes vascular ultrasound for physiological assessment of coronary
and atherosclerotic involvement of the coronary arteries of male stenosis: relationship between quantitative ultrasound tomography
and female subjects up to 40 years of age. Atherosclerosis 1981; and pressure-derived fractional flow reserve. Circulation 1999; 100:
38:39–50. 250–255.
11. Isner JM, Donaldson RF, Fortin AH, et al. Attenuation of the 30. Abizaid AS, Mintz GS, Mehran R, et al. Long-term follow-up
media of coronary arteries in advanced atherosclerosis. Am J after percutaneous transluminal coronary angioplasty was not
Cardiol 1986; 58:937–939. performed based on intravascular ultrasound findings: impor-
12. Topol EJ, Nissen SE. Our preoccupation with coronary luminol- tance of lumen dimensions. Circulation 1999; 100:256–261.
ogy. The dissociation between clinical and angiographic findings 31. Hermiller JB, Buller CE, Tenaglia AN, et al. Unrecognized left
in ischemic heart disease. Circulation 1995; 92:2333–2342. main coronary artery disease in patients undergoing interven-
13. Mintz GS, Painter JA, Pichard AD, et al. Atherosclerosis in tional procedures. Am J Cardiol 1993; 71:173–176.
angiographically “normal” coronary artery reference segments: 32. Gerber TC, Erbel R, Gorge G, et al. Extent of atherosclerosis and
an intravascular ultrasound study with clinical correlations. J Am remodeling of the left main coronary artery determined by
Coll Cardiol 1995; 25:1479–1485. intravascular ultrasound. Am J Cardiol 1994; 73:666–671.
[2/7/010/9:3:52] [355–369]
33. Ge J, Liu F, Gorge G, et al. Angiographically ‘silent’ plaque in the 51. Orford JL, Denktas AE, Williams BA, et al. Routine intravascular
left main coronary artery detected by intravascular ultrasound. ultrasound scanning guidance of coronary stenting is not associated
Coron Artery Dis 1995; 6:805–810. with improved clinical outcomes. Am Heart J 2004; 148: 501–506.
34. Abizaid AS, Mintz GS, Abizaid A, et al. One-year follow-up after 52. de Jaegere P, Mudra H, Figulla H, et al. Intravascular ultrasound-
intravascular ultrasound assessment of moderate left main cor- guided optimized stent deployment. Immediate and 6 months
onary artery disease in patients with ambiguous angiograms. J clinical and angiographic results from the Multicenter Ultra-
Am Coll Cardiol 1999; 34:707–715. sound Stenting in Coronaries Study (MUSIC Study). Eur Heart
35. Jasti V, Ivan E, Yalamanchili V, et al. Correlations between J 1998; 19:1214–1223.
fractional flow reserve and intravascular ultrasound in patients 53. Albiero R, Rau T, Schluter M, et al. Comparison of immediate
with an ambiguous left main coronary artery stenosis. Circula- and intermediate-term results of intravascular ultrasound versus
tion 2004; 110:2831–2836. angiography-guided Palmaz-Schatz stent implantation in
36. Isner JM, Kishel J, Kent KM, et al. Accuracy of angiographic matched lesions. Circulation 1997; 96:2997–3005.
determination of left main coronary arterial narrowing. 54. Gil RJ, Pawlowski T, Dudek D, et al. Comparison of angio-
Angiographic–histologic correlative analysis in 28 patients. Cir- graphically guided direct stenting technique with direct stenting
culation 1981; 63:1056–1064. and optimal balloon angioplasty guided with intravascular ultra-
37. Cameron A, Kemp HG Jr., Fisher LD, et al. Left main coronary sound. The multicenter, randomized trial results. Am Heart J
artery stenosis: angiographic determination. Circulation 1983; 68: 2007; 154:669–675.
484–489. 55. Sousa A, Abizaid A, Mintz GS, et al. The influence of intra-
38. Bech GJ, Droste H, Pijls NH, et al. Value of fractional flow reserve vascular ultrasound guidance on in-hospital outcomes after
in making decisions about bypass surgery for equivocal left main stent implantation: results from the Brazilian Society of Inter-
coronary artery disease. Heart 2001; 86:547–552. ventional Cardiology Registry—CENIC. J Am Coll Cardiol
39. Zhou Y, Kassab GS, Molloi S. On the design of the coronary 2002; 39:54A (abstr).
arterial tree: a generalization of Murray’s law. Phys Med Biol 56. Fitzgerald PJ, Oshima A, Hayase M, et al. Final results of the Can
1999; 44:2929–2945. Routine Ultrasound Influence Stent Expansion (CRUISE) study.
40. Maehara A, Mintz GS, Ahmed JM, et al. An intravascular ultra- Circulation 2000; 102:523–530.
sound classification of angiographic coronary artery aneurysms. 57. Frey AW, Hodgson JM, Müller C, et al. Ultrasound-guided
Am J Cardiol 2001; 88:365–370. strategy for provisional stenting with focal balloon combination
41. Ahmed JM, Mintz GS, Weissman NJ, et al. Mechanism of lumen catheter: results from the randomized Strategy for Intracoronary
enlargement during intracoronary stent implantation: an intra- Ultrasound-guided PTCA and Stenting (SIPS) trial. Circulation
vascular ultrasound study. Circulation 2000; 102:7–10. 2000; 102:2497–2502.
42. Maehara A, Takagi A, Okura H, et al. Longitudinal plaque 58. Russo RJ, Silva PD, Teirstein PS, et al. A randomized controlled
redistribution during stent expansion. Am J Cardiol 2000; 86: trial of angiography versus intravascular ultrasound-directed
1069–1072. bare-metal coronary stent placement (the AVID Trial). Circ
43. Prati F, Pawlowski T, Gil R, et al. Stenting of culprit lesions in Cardiovasc Interv 2009; 2:113–123.
unstable angina leads to a marked reduction in plaque burden: a 59. Gaster AL, Slothuus U, Larsen J, et al. Cost-effectiveness analysis
major role of plaque embolization? A serial intravascular ultra- of intravascular ultrasound guided percutaneous coronary inter-
sound study. Circulation 2003; 107:2320–2325. vention versus conventional percutaneous coronary intervention.
44. de Feyter PJ, Kay P, Disco C, et al. Reference chart derived from Scand Cardiovasc J 2001; 35:80–85.
post-stent-implantation intravascular ultrasound predictors of 6- 60. Schiele F, Meneveau N, Vuillemenot A, et al. Impact of intra-
month expected restenosis on quantitative coronary angiogra- vascular ultrasound guidance in stent deployment on 6-month
phy. Circulation 1999; 100:1777–1783. restenosis rate: a multicenter, randomized study comparing two
45. Choi JW, Goodreau LM, Davidson CJ. Resource utilization and strategies—with and without intravascular ultrasound guidance.
clinical outcomes of coronary stenting: a comparison of intra- J Am Coll Cardiol 1998; 32:320–328.
vascular ultrasound and angiographical guided stent implanta- 61. Schiele F, Meneveau N, Gilard M, et al. Intravascular ultrasound-
tion. Am Heart J 2001; 142:112–118. guided balloon angioplasty compared with stent: immediate and
46. Frey AW, Hodgson JM, Muller C, et al. Ultrasound-guided 6-month results of the multicenter, randomized Balloon Equiva-
strategy for provisional stenting with focal balloon combination lent to Stent Study (BEST). Circulation 2003; 107:545–551.
catheter: results from the randomized Strategy for Intracoronary 62. Cheneau E, Leborgne L, Mintz GS, et al. Predictors of subacute
Ultrasound-guided PTCA and Stenting (SIPS) trial. Circulation stent thrombosis: results of a systematic intravascular ultrasound
2000; 102:2497–2502. study. Circulation 2003; 108:43–47.
47. Gaster AL, Slothuus Skjoldborg U, Larsen J, et al. Continued 63. Fujii K, Carlier SG, Mintz GS, et al. Stent underexpansion and
improvement of clinical outcome and cost effectiveness follow- residual reference segment stenosis are related to stent throm-
ing intravascular ultrasound guided PCI: insights from a pro- bosis after sirolimus-eluting stent implantation: an intravascular
spective, randomised study. Heart 2003; 89:1043–1049. ultrasound study. J Am Coll Cardiol 2005; 45:995–998.
48. Schiele F, Meneveau N, Vuillemenot A, et al. Impact of intra- 64. Okabe T, Mintz GS, Buch AN, et al. Intravascular ultrasound
vascular ultrasound guidance in stent deployment on 6-month parameters associated with stent thrombosis after drug-eluting
restenosis rate: a multicenter, randomized study comparing two stent deployment. Am J Cardiol 2007; 100:615–620.
strategies—with and without intravascular ultrasound guidance. 65. Roy P, Steinberg DH, Sushinsky SJ, et al. The potential clinical
RESIST Study Group. REStenosis after Ivus guided STenting. utility of intravascular ultrasound guidance in patients undergo-
J Am Coll Cardiol 1998; 32:320–328. ing percutaneous coronary intervention with drug-eluting stents.
49. Oemrawsingh PV, Mintz GS, Schalij MJ, et al. Intravascular Eur Heart J 2008; 29:1851–1857.
ultrasound guidance improves angiographic and clinical out- 66. Maehara A, Mintz GS, Bui AB, et al. Incidence, morphology,
come of stent implantation for long coronary artery stenoses: angiographic findings, and outcomes of intramural hematomas
final results of a randomized comparison with angiographic after percutaneous coronary interventions: an intravascular
guidance (TULIP Study). Circulation 2003; 107:62–67. ultrasound study. Circulation 2002; 105:2037–2042.
50. Mudra H, di Mario C, de Jaegere P, et al. Randomized compar- 67. Hoffmann R, Mintz GS, Pichard AD, et al. Intimal hyperplasia
ison of coronary stent implantation under ultrasound or angio- thickness at follow-up is independent of stent size: a serial
graphic guidance to reduce stent restenosis (OPTICUS Study). intravascular ultrasound study. Am J Cardiol 1998; 82:
Circulation 2001; 104:1343–1349. 1168–1172.
[2/7/010/9:3:52] [355–369]
68. Moses JW, Leon MB, Popma JJ, et al. Sirolimus-eluting stents 86. Kaaresen KF, Bolviken E. Blind deconvolution of ultrasonic
versus standard stents in patients with stenosis in a native traces accounting for pulse variance. IEEE Trans Ultrason Fer-
coronary artery. N Engl J Med 2003; 349:1315–1323. roelectr Freq Control 1999; 46:564–573.
69. Sousa JE, Costa MA, Sousa AG, et al. Two-year angiographic and 87. Konig A, Klauss V. Virtual histology. Heart 2007; 93:977–982.
intravascular ultrasound follow-up after implantation of siroli- 88. Nasu K, Tsuchikane E, Katoh O, et al. Accuracy of in vivo
mus-eluting stents in human coronary arteries. Circulation 2003; coronary plaque morphology assessment: a validation study of
107:381–383. in vivo virtual histology compared with in vitro histopathology.
70. Stone GW, Midei M, Newman W, et al. Comparison of an J Am Coll Cardiol 2006; 47:2405–2412.
everolimus-eluting stent and a paclitaxel-eluting stent in patients 89. Nair A, Margolis P, Kuban BD, et al. Automated coronary plaque
with coronary artery disease: a randomized trial. JAMA 2008; characterization with intravascular ultrasound backscatter: ex
299:1903–1913. vivo validation. EuroIntervention 2007; 3:113–120.
71. Weissman NJ, Koglin J, Cox DA, et al. Polymer-based paclitaxel- 90. Virmani R, Kolodgie FD, Burke AP, et al. Lessons from sudden
eluting stents reduce in-stent neointimal tissue proliferation: a coronary death: a comprehensive morphological classification
serial volumetric intravascular ultrasound analysis from the scheme for atherosclerotic lesions. Arterioscler Thromb Vasc
TAXUS-IV trial. J Am Coll Cardiol 2005; 45:1201–1205. Biol 2000; 20:1262–1275.
72. Miyazawa A, Ako J, Hongo Y, et al. Comparison of vascular 91. Kolodgie FD, Burke AP, Farb A, et al. The thin-cap fibroather-
response to zotarolimus-eluting stent versus sirolimus-eluting oma: a type of vulnerable plaque: the major precursor lesion to
stent: intravascular ultrasound results from ENDEAVOR III. Am acute coronary syndromes. Curr Opin Cardiol 2001; 16:285–292.
Heart J 2008; 155:108–113. 92. Burke AP, Farb A, Malcom GT, et al. Coronary risk factors and
73. Ormiston JA, Serruys PW, Regar E, et al. A bioabsorbable ever- plaque morphology in men with coronary disease who died
olimus-eluting coronary stent system for patients with single suddenly. N Engl J Med 1997; 336:1276–1282.
de-novo coronary artery lesions (ABSORB): a prospective open- 93. Fishbein MC, Siegel RJ. How big are coronary atherosclerotic
label trial. Lancet 2008; 371:899–907. plaques that rupture? Circulation 1996; 94:2662–2666.
74. Castagna MT, Mintz GS, Leiboff BO, et al. The contribution of 94. Schaar JA, De Korte CL, Mastik F, et al. Characterizing vulner-
“mechanical” problems to in-stent restenosis: An intravascular able plaque features with intravascular elastography. Circulation
ultrasonographic analysis of 1090 consecutive in-stent restenosis 2003; 108:2636–2641.
lesions. Am Heart J 2001; 142:970–974. 95. Hong MK, Mintz GS, Lee CW, et al. A three-vessel virtual histology
75. Mintz GS, Weissman NJ. Intravascular ultrasound in the drug- intravascular ultrasound analysis of frequency and distribution of
eluting stent era. J Am Coll Cardiol 2006; 48:421–429. thin-cap fibroatheromas in patients with acute coronary syndrome
76. Hong MK, Mintz GS, Lee CW, et al. Incidence, mechanism, or stable angina pectoris. Am J Cardiol 2008; 101:568–572.
predictors, and long-term prognosis of late stent malapposition 96. Missel E, Mintz GS, Carlier SG, et al. Necrotic core and its ratio to
after bare-metal stent implantation. Circulation 2004; 109: dense calcium are predictors of high-risk non-ST-elevation acute
881–886. coronary syndrome. Am J Cardiol 2008; 101:573–578.
77. Serruys PW, Degertekin M, Tanabe K, et al. Intravascular ultra- 97. Kawaguchi R, Oshima S, Jingu M, et al. Usefulness of virtual
sound findings in the multicenter, randomized, double-blind histology intravascular ultrasound to predict distal embolization
RAVEL (RAndomized study with the sirolimus-eluting VElocity for ST-segment elevation myocardial infarction. J Am Coll Car-
balloon-expandable stent in the treatment of patients with de novo diol 2007; 50:1641–1646.
native coronary artery Lesions) trial. Circulation 2002; 106:798–803. 98. Uetani T, Amano T, Ando H, et al. The correlation between lipid
78. Tanabe K, Serruys PW, Degertekin M, et al. Incomplete stent volume in the target lesion, measured by integrated backscatter
apposition after implantation of paclitaxel-eluting stents or bare intravascular ultrasound, and post-procedural myocardial infarc-
metal stents: insights from the randomized TAXUS II trial. Cir- tion in patients with elective stent implantation. Eur Heart J 2008;
culation 2005; 111:900–905. 29:1714–1720.
79. Siqueira DA, Abizaid AA, Costa Jde R, et al. Late incomplete 99. Rodriguez-Granillo GA, Serruys PW, Garcia-Garcia HM, et al.
apposition after drug-eluting stent implantation: incidence and Coronary artery remodelling is related to plaque composition.
potential for adverse clinical outcomes. Eur Heart J 2007; 28: Heart 2006; 92:388–391.
1304–1309. 100. Surmely JF, Nasu K, Fujita H, et al. Association of coronary
80. Hoffmann R, Morice MC, Moses JW, et al. Impact of late incom- plaque composition and arterial remodelling: a virtual histology
plete stent apposition after sirolimus-eluting stent implantation analysis by intravascular ultrasound. Heart 2007; 93:928–932.
on 4-year clinical events: intravascular ultrasound analysis from 101. Fujii K, Carlier SG, Mintz GS, et al. Association of plaque char-
the multicentre, randomised, RAVEL, E-SIRIUS and SIRIUS acterization by intravascular ultrasound virtual histology and
trials. Heart 2008; 94:322–328. arterial remodeling. Am J Cardiol 2005; 96:1476–1483.
81. Joner M, Finn AV, Farb A, et al. Pathology of drug-eluting stents 102. Burke AP, Kolodgie FD, Farb A, et al. Morphological predictors
in humans: delayed healing and late thrombotic risk. J Am Coll of arterial remodeling in coronary atherosclerosis. Circulation
Cardiol 2006; 48:193–202. 2002; 105:297–303.
82. Cook S, Wenaweser P, Togni M, et al. Incomplete stent apposi- 103. Granada JF, Wallace-Bradley D, Win HK, et al. In vivo plaque
tion and very late stent thrombosis after drug-eluting stent characterization using intravascular ultrasound-virtual histology
implantation. Circulation 2007; 115:2426–2434. in a porcine model of complex coronary lesions. Arterioscler
83. Rodriguez-Granillo GA, Garcia-Garcia HM, Mc Fadden EP, et al. Thromb Vasc Biol 2007; 27:387–393.
In vivo intravascular ultrasound-derived thin-cap fibroatheroma 104. Kawasaki M, Sano K, Okubo M, et al. Volumetric quantitative
detection using ultrasound radiofrequency data analysis. J Am analysis of tissue characteristics of coronary plaques after statin
Coll Cardiol 2005; 46:2038–2042. therapy using three-dimensional integrated backscatter intravas-
84. Palmer ND, Northridge D, Lessells A, et al. In vitro analysis of cular ultrasound. J Am Coll Cardiol 2005; 45:1946–1953.
coronary atheromatous lesions by intravascular ultrasound; 105. Mercado N, Maier W, Boersma E, et al. Clinical and angiographic
reproducibility and histological correlation of lesion morphol- outcome of patients with mild coronary lesions treated with
ogy. Eur Heart J 1999; 20:1701–1706. balloon angioplasty or coronary stenting. Implications for
85. Gaston-Rodrigues GA, Serruys PW. OCT plaque characterization: mechanical plaque sealing. Eur Heart J 2003; 24:541–551.
comparison to IVUS-VH. In: Regar E, van Leeuwen TG, Serruys 106. Nikolsky E, Mehran R, Mintz GS, et al. Drug-Eluting Stents in the
PW, eds. Optical Coherence Tomography in Cardiovascular Treatment of Intermediate Lesions: Pooled Analysis from 14
Research. London: Informa Healthcare, 2007. Randomized Clinical Trials. Circulation 2008; 118:S_650.
[2/7/010/9:6:4] [370–387]
29
Optical coherence tomography

E. Regar, N. Gonzalo, G. Van Soest, and P. W. Serruys
INTRODUCTION volume to the total cell volume in a manner similar to what a

Optical coherence tomography (OCT) is a light-based imaging pathologist does when diagnosing cancer. In the days since the
modality showing tremendous potential in the coronary circu- initial discussion in Prof Fujimoto’s office, several members of
lation. Compared with traditional intravascular ultrasound the MIT OCT team have started academic OCT research pro-
(IVUS), OCT has a 10-fold higher image resolution resulting grams throughout the United States.
from use of light rather than sound. OCT has been successfully Compared with an OCT microscope, used in ophthal-
applied to the assessment of atherosclerotic plaque, stent appo- mology and in most experimental settings, the application of
sition, and tissue coverage, introducing a new era in intra- OCT within the human vascular system, particularly within
vascular coronary imaging. coronary arteries, represents a challenge, as a number of prin-
The origins of OCT date back to 1990, when Dr David cipal problems need to be overcome. Hence, the intracoronary
Huang was in his fourth year of an MD-PhD program at application of OCT has developed relatively slowly, but stead-
Massachusetts Institute of Technology (MIT). As an offshoot ily, over the last decade, and a commercially available system
of ongoing femtosecond ranging projects in Prof James Fujimo- for clinical use (LightLab Imaging, Inc., Westford, Massachu-
to’s laboratory, Dr Huang had been studying optical coherence setts, U.S.) is being approved in Europe and Japan. Today, the
domain reflectometry (OCDR) to perform ranging measure- technology development from time-domain OCT to Fourier-
ments in the eye. The retinal scans, however, were very hard to domain OCT has the potential to dramatically change the
interpret. The thought occurred to Dr Huang that adding research landscape allowing for a widespread clinical intracoro-
transverse scanning to OCDR graphs would create an image nary application in research and patient care. This chapter
that would be much easier for a human to interpret than the set discusses these technical principles of intracoronary OCT,
of OCDR waveforms. All that was required was to add a summarizes the preclinical and clinical research, discusses
translation stage and a software package to convert a data potential clinical applications, and explains the practical per-
matrix into an image. formance of OCT in the catheterization laboratory. Differences
The central problem in making tomographic images in time-domain versus Fourier-domain OCT will be pointed out
using light was to develop a technique that would permit whenever relevant.
reflections from various depths to be measured and recorded
in a fashion analogous to ultrasonic imaging. In the case of
sound, electronic circuits are fast enough to separate the echoes ANATOMIC CONSIDERATIONS
from structures that are within the resolution cell of the ultra- Principally, all epicardial coronary arteries, venous or arterial
sonic transducer. In the case of light, an interferometer has to be grafts accessible by a guiding catheter, are eligible for OCT
employed to overcome measurement difficulties caused by the imaging.
speed of light, which is much faster than the speed of sound. By Considerations regarding anatomy and patient character-
using an interferometer it was, for the first time, possible to istics arise (i) from the fact that OCT imaging requires a blood-
record reflections from various depths in a biological tissue. free environment and (ii) from OCT catheter design. As the
Since 1990, OCT technology has generated over 5000 imaging procedure demands temporary blood removal and
articles in academic journals. The first manuscript from MIT, flush (e.g., lactated ringers or X-ray contrast medium), it should
published in 1991, describes the basic concept of an OCT not be performed in patients with severely impaired left
imaging system and discusses its possible applications in ventricular function or those presenting with hemodynamic
both retinal and arterial imaging (1). In 1996, a second manu- compromise. Further, OCT should be used with caution in
script was published that dealt specifically with the possibility patients with single remaining vessel or those with markedly
of imaging coronary arteries with an OCT device (2). impaired renal function. Lesions that are ostially or proximally
It became clear early on that OCT could contribute to the located cannot be adequately imaged using proximal balloon
diagnosis of ocular diseases. It was believed that the new occlusion and, thus, a nonocclusive technique may be preferred
technology had the potential to serve as an in vivo microscope in these circumstances. Large caliber vessels or very tortuous
that could obtain nonexcisional biopsy information from loca- vessels often preclude complete circumferential imaging as a
tions at which a conventional biopsy was either impossible or result of a noncentral, noncoaxial position of the OCT imaging
impractical to perform. A second research thrust from the MIT probe within the vessel.
group was to push the resolution of the technology to increas- These anatomic limitations seem to be significantly atte-
ingly higher levels using wider bandwidth optical sources. nuated in Fourier-domain OCT as the pullback speed is much
Sufficiently wide bandwidth sources would enable to resolve higher, and as a result, the duration of ischemia and the
subcellular structures and measure the ratio of the nuclear amount of potentially nephrotoxic flush is much lower.
[2/7/010/9:6:4] [370–387]
OPTICAL COHERENCE TOMOGRAPHY 371
Table 29.1 Comparison of OCT and IVUS

OCT IVUS
Image example
Axial resolution 10–20 mm 100–150 mm

Penetration depth into tissue 1.5–2 mm 4–8 mm
Probe size 0.4 mm 1.1 mm
Pullback speed Up to 40 mm/sec 1.0 mm/sec
Blood removal needed Yes No
Plaque characterization Yes Yes
Fibrous cap measurement Yes No
Vessel remodeling No Yes
Increased penetration depth and scanning range allow imaging mainly determined by the imaging optics in the catheter and is
of the complete circumference of large and tortuous vessels. approximately 25 mm. The imaging depth of approximately 1.0
The design as short monorail catheter enables to negotiate even to 1.5 mm within the coronary artery wall is limited by the
complex lesions by selecting an appropriate standard guide- attenuation of light in the tissue (Table 29.1). Analogous to
wire. As there is no proximal balloon occlusion necessary, also ultrasound imaging, the echo time delay of the emitted light is
ostial lesions, bifurcations and large vessels can be visualized. used to generate spatial image information, the intensity of the
received (reflected or scattered) light is translated into a (false)
color scale. As the speed of light is much faster than that of
sound, an interferometer is required to measure the backscat-
FUNDAMENTALS
tered light (3). The interferometer splits the light source into
The principle is analogous to pulse-echo ultrasound imaging;
two “arms”—a reference arm and a sample arm, which is
however, light is used rather than sound to create the image.
directed into the tissue. The light from both arms is recombined
Although ultrasound produces images from backscattered
at a detector, which registers the so-called interferogram, the
sound “echoes,” OCT uses infrared light waves that reflect
sum of reference and sample arm fields. Because of the large
off the internal microstructure within the biological tissues. The
source bandwidth, the interferogram is nonzero only if the
use of light allows for a 10-fold higher image resolution; how-
sample and reference arms are of equal length, within a small
ever, this is at the expense of a reduced penetration depth and
window equal to the coherence length of the light source (4,5).
the need to create a blood-free environment for imaging. In
coronary arteries, blood (namely, red blood cells) represents
that nontransparent tissue causing multiple scattering and
substantial signal attenuation. As a consequence, blood must Time-Domain OCT
be displaced during OCT imaging. This procedure is poten- In time-domain OCT, the length of the reference arm is scanned
tially causing ischemia in the territory of the artery under over a distance of typically a few millimeters, by moving a
study. The need for balloon occlusion and intracoronary flush mirror. The point from which intensity is collected from the
are at the forefront of emerging developments to simplify the sample arm is moved through the tissue accordingly, and the
OCT image acquisition process. Automated catheter pullbacks amplitude of the recorded interferogram in a scan corresponds
at very high speed are currently under development in OCT to the reflectivity of the tissue along the direction of the sample
systems using optical Fourier-domain imaging. Faster pullback beam. By scanning the beam along the tissue, in a rotary fashion
speeds offer the potential to scan an entire stent within a matter for intravascular imaging, an image is built up out of neighbor-
of five to six seconds. ing lines. Figure 29.1 shows the currently commercially avail-
able time-domain OCT system (LightLab Imaging, Inc.).
OCT Principles
OCT utilizes a near infrared light source (1300-nm wave- Fourier-Domain OCT
length) in combination with advanced fiber-optics to create a A new generation of OCT systems operates in the frequency
dataset of the coronary artery. Both the bandwidth of the (rather than time) domain, also called Fourier domain. The
infrared light used and the wave velocity are orders of magni- interferogram is detected as a function of wavelength, either by
tude higher than in medical ultrasound. The resulting resolu- using a broadband source as in the time domain systems, and
tion depends primarily on the ratio of these parameters, and is spectrally resolved detection, or alternatively by incorporating
one order of magnitude higher than that of IVUS: the axial a novel wavelength-swept laser source (6,7) (Table 29.2). This
resolution of OCT is about 15 mm. The lateral resolution is latter technique is also called “swept-source OCT,” or optical
[2/7/010/9:6:4] [370–387]
Figure 29.1 The currently commercially available TD-OCT

system (LightLab Imaging, Inc., Westford, Massachusetts,
U.S.). (A) The OCT imaging wire with an outer diameter of
0.019 in.; (B) magnification of the distal catheter tip;
(C) magnification of the 0.006-in. rotating single-mode
fiber-optic core, located within the distal sleeve of the
imaging wire; (D) imaging console with pullback device
allowing for real-time image display and data storage.
Abbreviation: OCT, optical coherence tomography.
Table 29.2 Characteristics of TD-OCT and FD-OCT

TD-OCT FD-OCT
Fixed light source and variable reference Light source with variable wavelength and
mirror fixed reference mirror
Diagram
Axial resolution 10–20 mm Up to 7 mm

Penetration depth 1.5–2 mm 1.5–2 mm
Optical core ImageWire Integrated catheter
Maximum pullback speed 3 mm/sec Up to 40 mm/sec
Scan diameter 7 mm >10 mm
Blood removal needed Yes Yes
Characteristics of TD-OCT are given for the commercially available LightLab Imaging, Inc. (Westford, Massachusetts, U.S.) system,
characteristics of FD-OCT are based on noncommercially available prototypes.
Abbreviations: TD-OCT, time-domain optical coherence tomography; FD-OCT, Fourier-domain optical coherence tomography.
frequency domain imaging (OFDI), and capitalizes most effec- order of 105 lines/sec. In a Fourier-domain OCT system, the
tively on the higher sensitivity and signal-to-noise ratio offered wavelength range of the sweep determines the resolution of the
by Fourier-domain detection. This development has led to image, while the imaging depth is inversely related to the
faster image acquisition speeds, with greater penetration instantaneous spectral width of the source.
depth, without loss of vital detail or resolution, and represents The increased sensitivity of Fourier-domain OCT also
a great advancement on current conventional OCT systems. allows for larger imaging depths. The attenuation of light by
From the signal received in one wavelength sweep, the depth the tissue is the same for time-domain and for Fourier-domain
profile can be constructed by the Fourier transform operation OCT, but the lower noise of the latter makes it possible to
that is performed electronically in the data processing unit. All discern weaker signals that would be indistinguishable from
other components of a Fourier-domain system (the interferom- the background in time-domain OCT. The depth range from
eter, the catheter, including the imaging optics, display) are which useful anatomical information can be extracted is
comparable in principle to those in a time-domain OCT system. extended by a factor of about 3 (12). Clinically, this advantage
The scan speed, or line rate, in a time-domain OCT enables the assessment of coronary microstructures well
system are limited by the achievable mechanical scan speed beyond the arterial-lumen border.
of the reference arm mirror and by the sensitivity of the signal Fourier-domain OCT systems produce images much faster
detection (8). The source wavelength in Fourier-domain OCT than standard video-rate, so recorded data has to be replayed for
can be swept at a much higher rate than the position scan of the inspection by the operator. Currently, OCT systems scan 200 to
reference arm mirror in a time-domain OCT system. In addi- 500 angles per revolution (frame), and 5 to 10 images/mm in a
tion, Fourier-domain OCT has a higher sensitivity than time- pullback. If these parameters are maintained with high-speed
domain OCT at large line rates and scan depths (9–11). These systems, 20 mm/sec (or higher) pullback speeds are possible at
features can be put to good use in larger scan speeds, of the the same sampling density as conventional OCT data. Figure 29.2
[2/7/010/9:6:4] [370–387]
Figure 29.2 Different FD-OCT prototypes as used at the Thoraxcenter in 2008. (A) M4 system (LightLab Imaging, Inc., Westford,
Massachusetts, U.S.), (B) Terumo OCT, (C) Volcano OCT, (D) MGH OFDI system. Abbreviations: FD-OCT, Fourier-domain optical
coherence tomography; OFDI, optical frequency domain imaging; MGH OFDI, Massachusetts general Hospital optical frequency domain
imaging. Source: Courtesy of G. Tearney and B. Bouma, Wellman Center for Photomedicine, MGH, Boston, Massachusetts, U.S.A.
Figure 29.3 FD-OCT images as obtained in in vivo in normal porcine coronary arteries. The coronary vessel wall shows a three-layer
appearance, and an intimal dissection is visible in the 4 o’clock position. (A) M4 prototype (LightLab Imaging, Inc., Westford, Massachusetts,
U.S.), (B) Terumo OCT, (C) MGH OFDI system. Abbreviations: FD-OCT, Fourier-domain optical coherence tomography; OFDI, optical
frequency domain imaging; MGH OFDI, Massachusetts general Hospital optical frequency domain imaging.
shows different Fourier-domain OCT prototypes as used at the scan is limited. The high data rate of novel OCT technologies
Thoraxcenter in 2008. Figure 29.3 illustrates Fourier-domain OCT could also be used to increase sampling density either in the
images as obtained in vivo in normal porcine coronary arteries. longitudinal (pullback) or angular direction. A smaller spacing
The high scan speeds have been employed for real-time between frames in a pullback would lead to a better sampling
volumetric imaging of dynamic phenomena including fast of small-scale features in the arterial or stent geometry that
pullbacks for intracoronary imaging with minimal ischemia would be missed at 100 mm inter-frame distance. Denser sam-
(12), and retinal scans with minimal motion artifacts (13). pling in the angular direction would facilitate speckle filtering
Imaging of dynamic phenomena in time, or rather removing in OCT images. Speckle is a major obstacle for the development
motion artifacts, is the prime application of high-speed OCT. of parametric and quantitative imaging techniques. These
3D rendering of volumes becomes possible if motion during the possibilities are still largely unexplored.
[2/7/010/9:6:4] [370–387]
Figure 29.4 OCT findings before and immediately after stent implantation in a patient presenting with acute myocardial infarction (M4
prototype FD-OCT system, LightLab Imaging, Inc., Westford, Massachusetts, U.S.A.). (A) Coronary angiography revealed a subtotal
occlusion of the left anterior descending artery. The black arrow indicates the vessel segment imaged by OCT. Preinterventional OCT shows
(B) the distal reference, (C) the lesion with mural thrombus, (D) the minimal lumen area, and (E) the proximal reference. Corresponding
postinterventional OCT images can be easily identified using anatomical landmarks such as side branches (SB) or calcium nodules. Metallic
stent struts appear as bright structures with dorsal shadowing. The distal reference (F) and the lesion site that can be easily identified using
the small calcium nodule as a landmark (G), the site of the original minimal lumen area shows tissue protrusion in the 6 o’clock position (H),
the proximal reference (I). Abbreviation: FD-OCT, Fourier-domain optical coherence tomography.
INDICATIONS phenomenon being particularly evident in regions of stent

To date, there is no established indication for the clinical use of overlap. In an evaluation of OCT findings following stent
OCT in patient care. However, there are a variety of research implantation to complex coronary lesions, Tanigawa et al.
applications where OCT is clearly accepted as the “gold (18) examined a total of 6402 struts from 23 patients (25 lesions)
standard.” We summarize the accepted research applications and found 9.1 7.4% of all struts in each lesion treated were
and illustrate the potential clinical implications of OCT findings. malapposed. Univariate predictors of malapposition on multi-
level logistic regression analysis were the implantation of a
sirolimus-eluting stent (SES, likely due to its increased strut
The Role of OCT in Stent Imaging thickness and closed-cell design) together with the presence of
Coronary artery lesions and results following percutaneous
overlapping stents, a longer stent length, and a type C lesion.
coronary intervention (PCI) are usually assessed angiographi-
Likely mechanical explanations for stent malapposition include
cally. This luminogram technique provides a unique overview
strut thickness, closed-cell design, or acute stent recoil (19).
of the coronary tree, gives information regarding anatomy and
Although these findings are impressive and helpful for the
topography, and can confirm the presence of atherosclerosis
improvement of future stent designs, today the clinical rele-
with high specificity. The prognostic relevance for subsequent
vance and potentially long-term sequelae of malapposed struts
cardiac events, such as myocardial infarction, however, is lim-
as detected by OCT are currently unknown. Figure 29.4 illus-
ited. Furthermore, stent implantation and optimization under-
trates OCT findings before and immediately after stent implan-
taken using angiographic guidance alone has been shown to
tation in a patient presenting with acute myocardial infarction.
result in more frequent incomplete stent expansion and an
Figure 29.5 illustrates incomplete stent strut apposition as
increased future risk of target vessel revascularization when
frequently seen by OCT immediately after stenting.
compared with guidance with IVUS.
Assessment of Acute Stent Apposition Assessment of Long-Term Outcome

For the past two decades, IVUS has been used to assess the Visualization and quantification of stent strut tissue
acute result following stenting, giving valuable information on coverage OCT can reliably detect early and very thin layers
stent expansion, strut apposition, and signs of vessel trauma of tissue coverage on stent struts. Figure 29.6 illustrates the
including dissections and tissue prolapse. IVUS studies (14,15) typical OCT appearance of neointima in bare-metal stents
suggested that stent strut malapposition is a relatively uncom- (BMS) and in drug-eluting stents (DES). Several small studies
mon finding, observed in approximately 7% of cases, and that have recently been published highlighting the application of
strut malapposition does not increase the risk of subsequent OCT in the detection of stent tissue coverage at follow-up.
major adverse cardiac events. In contrast, OCT can visualize the Importantly, OCT permits the quantification of tissue coverage
complex coronary arterial wall structure after stenting in much with high reliability (20). Matsumoto et al. (21) studied
greater detail (16). As a result, OCT studies in the acute 34 patients following SES implantation. The mean neointima
poststent setting (17) have demonstrated a relatively high pro- thickness was 52.5 mm, and the prevalence of struts covered by
portion of stent struts, not completely apposed to the vessel thin neointima undetectable by IVUS was 64%. The average
wall contact, even after high-pressure postdilatation with this rate of neointima-covered struts in an individual SES was 89%.
[2/7/010/9:6:4] [370–387]
three-month group (5% vs. 15%, respectively; p < 0.001). Con-

versely, prevalence of patients with uncovered struts did not
differ between the three-month and the two-year follow-up
study (95% vs. 81%, respectively), highlighting that exposed
struts continued to persist at long-term follow-up. Chen et al.
(24) recently used OCT to image SES and BMS at different time
points following implantation. Of the 10 SES and 13 BMS
imaged, the authors identified a significantly higher number
of incompletely apposed and uncovered stent struts in patients
receiving SES compared with BMS. Figure 29.7 illustrates OCT
finding at long-term follow-up after DES implantation.
Assessment of structural details of tissue coverage OCT
also permits the qualitative characterization of neointimal tis-
sue in a qualitative way (25). This is a great advantage as such
information has not been available in vivo until now. The
limited resolution together with artifacts induced by metallic
stent struts do not allow the characterization of such details by
IVUS. With OCT, neointimal tissue can show a variety of
morphologies ranging from homogenous, bright, uniform tis-
sue to optically heterogeneous tissue or eccentric tissue of
various thickness. Furthermore, structural details within the
Figure 29.5 Incomplete stent strut apposition immediately after tissue can be observed such as intimal neovascularization (26)
stenting (Volcano FD-OCT prototype). (A) Coronary angiogram of or a layered appearance (27), typically observed in restenotic
the right coronary artery after stent implantation, (B) OCT image of regions. Variations in the appearance of strut coverage can be
the proximal stent edge showing malapposition of stent struts, and
seen within an individual patient, within an individual stent, or
(C) magnification clearly illustrating that individual stent struts do not
within stents of different design.
have contact to the vessel wall. Abbreviation: FD-OCT, Fourier-
domain optical coherence tomography. OCT findings, such as dark, signal-poor halos around
stent struts may reflect fibrin deposition and incomplete heal-
ing, as described in pathological and animal experimental
series (28). However, there is paucity of data demonstrating
directly the OCT appearance of different components in neo-
Nine SES (16%) showed full coverage by neointima, whereas intimal tissue as defined by histology. Postmortem imaging of
the remaining stents had partially uncovered struts. Similarly, DES in human coronaries is difficult and might be limited by
Takano et al. (22) studied 21 patients (4516 struts) three months the fact that the optical tissue properties show variations with
following SES implantation. Rates of exposed struts and temperature and fixation (29). Long-term animal OCT observa-
exposed struts with malapposition were 15% and 6%, respec- tions in DES are scarce.
tively. These were more frequent in patients with acute coro-
nary syndrome (ACS) than in those with non-ACS (18% vs. OCT Assessment of Innovative Stent Designs and Materials
13%, p < 0.001; 8% vs. 5%, p < 0.005, respectively). The same OCT is also becoming an integral tool to assess emerging stent
group have recently reported two-year follow-up OCT findings technologies that are increasingly becoming more sophisticated
(23), with the thickness of neointimal tissue at two years being (e.g., bioabsorbable polymers and stents, biodegradable mag-
greater than that at three months (71 93 mm vs. 29 41 mm, nesium alloy stents) and thus demanding highly detailed
respectively; p < 0.001). Frequency of uncovered struts was assessments both in the initial animal testing phases and in
found to be lower in the two-year group compared with the the clinical trial setting. Morphological changes of the
Figure 29.6 In vivo OCT of human coronaries at six

months follow-up. This figure illustrates the different
degree of neointimal coverage typically seen in bare-
metal stents as compared with drug-eluting stents.
(A) bare-metal stent and (B) drug-eluting stent (TD-
OCT system, LightLab Imaging, Inc., Westford, Mas-
sachusetts, U.S.). Abbreviations: OCT, optical coher-
ence tomography; TD-OCT, time domain optical
coherence tomography.
[2/7/010/9:6:4] [370–387]
Figure 29.7 OCT findings at long-term follow-up after DES implantation in the LEADERS OCT substudy. (A) Coronary angiogram of the left
coronary artery. Overlapping stents were implanted in the proximal left anterior descending artery and in the left main stem. (B) The
application of the nonocclusive imaging acquisition technique allowed to image the stent in the left main stem as well as the proximal left
anterior descending artery. (C) Eccentric stent struts with very thin coverage visible, an individual strut in 11 o’clock position is malapposed.
(D) Stent struts showing malapposition and tissue coverage with irregular borders, possibly indicating thrombus formation (TD-OCT system,
LightLab Imaging, Inc., Westford, Massachusetts, U.S.). Abbreviations: DES, drug-eluting stent; OCT, optical coherence tomography;
TD-OCT, time domain optical coherence tomography.
absorbable, polylactic acid stent struts and the vessel wall reason for this interest is the fact that ACSs caused by the
during follow-up have been recently described and show the rupture of a coronary plaque are common initial and often fatal
unique capabilities of this in vivo imaging modality (30). The manifestations of coronary atherosclerosis in otherwise healthy
ABSORB trial recently published showed the feasibility of subjects. The detection of the lesions with high risk of rupture
implantation of the bioabsorbable everolimus-eluting coronary (the so-called “vulnerable plaques”) would be of main impor-
stent (BVS; Abbott Laboratories, Illinois, U.S.), composed of a tance for the prevention of future ACS. In the last years there
poly-L-lactic acid backbone, coated with a degradable polymer- has been a growing interest in this field, and a lot of different
everolimus matrix. OCT allowed not only a precise character- techniques have been developed to evaluate diverse aspects
ization of the stent apposition and coverage but also demon- involved in plaque vulnerability. Among them, OCT has
strated structural changes in the bioabsorbable DES over time. emerged as one of the most promising due to its ability to
At present OCT appears as the best available tool for the provide unique information about the plaque composition, the
assessment of the absorption of the stent struts. thickness of the fibrous cap, the presence of macrophages and
Another field of innovation consists in the treatment of tissue collagen composition.
bifurcation lesions with a variety of dedicated stents under
clinical investigation. Key issues in bifurcation stenting include Plaque Composition
the ability of a stent to cover different vessel calibers at the The propensity of atherosclerotic lesions to destabilize and
proximal lesion site, the carina, and the distal lesion site, and to rupture is highly dependent on their composition. In compar-
provide maximal expansion of the carina on both sides, the ison with histology, OCT has demonstrated to be highly sensi-
main vessel and the side branch. OCT can be a very useful tool tive and specific for characterizing different types of
to study the stent—vessel wall interaction as well as patency of atherosclerotic plaques (Figs. 29.12 and 29.13). Yabushita et al.
the carina in this more complex anatomy. Clinical and exper- (31) performed an in vitro study of more than 300 human
imental examples of OCT findings in bifurcation stenting are atherosclerotic artery segments. When compared with histolog-
given in Figures 29.8 to 29.11 and Table 29.3. ical examination, OCT had a sensitivity and specificity of 71%
to 79% and 97% to 98% for fibrous plaques, 95% to 96% and
Atherosclerotic Plaque Assessment 97% for fibrocalcific plaques, and 90% to 94.5% and 90% to 92%
OCT is highly sensitive and specific for the characterization of for lipid-rich plaques, respectively. Further, the interobserver
plaques when compared with histological examination. and intraobserver variability of OCT measurements were high
Recently, pathophysiology and coronary morphology in (k values of 0.88 and 0.91, respectively). Ex vivo validations
patients with ACS are getting more and more attention. One have also shown that OCT is superior to conventional and
[2/7/010/9:6:4] [370–387]
Figure 29.8 Clinical OCT after bifurcation stenting of the ramus descendens anterior and the ramus diagonalis in crush technique. (A) OCT
pullback is started distally in the diagonal branch. The stent is well expanded and all struts are apposed against the vessel wall. (B) OCT at the
carina shows the lumen of the ramus descendens anterior in 9 o’clock position wide open. Note the double layer of stent struts separation the
ostium of the diagonal branch from the ramus descendens anterior. (C) Proximal stent portion in the ramus descendens anterior. In 12 to 6
o’clock position, there is a layer of struts originating from the proximal, crushed diagonal branch stent, and from the stent in the ramus
descendens anterior can be observed (TD-OCT system, LightLab Imaging, Inc., Westford, Massachusetts, U.S.). Abbreviations: OCT, optical
coherence tomography; LAD, left anterior descending artery; RD, ramus diagonalis; TD-OCT, time domain optical coherence tomography.
Figure 29.9 Clinical OCT after bifurcation stenting of the ramus descendens anterior and the ramus diagonalis with a dedicated bifurcation
stent (Nile Croco stent, Minvasys, France). (A) OCT pullback is started distally in the ramus descendens anterior. The stent is well expanded
and all struts are apposed against the vessel wall. (B) OCT at the carina shows a wide open lumen with no stent struts obstructing the access
to the ramus diagonalis. (C) Proximal stent portion in the ramus descendens anterior. Stent strut malapposition is visible in 11 to 1 o’clock
position. (D) Longitudinal view from distal (left) to proximal (right). (FD-OCT prototype and M4 prototype FD-OCT system, LightLab Imaging,
Inc., Westford, Massachusetts, U.S.). Abbreviations: GW, guidewire; OCT, optical coherence tomography; FD-OCT, Fourier domain optical
coherence tomography.
integrated backscatter IVUS for the characterization of coronary (38) and men or women with ACS (39). According to histolog-
atherosclerotic plaque composition (32–35). In vivo, OCT is able ical and IVUS examinations, the percentage of lipid-rich plaque
to identify most of the architectural features identified by IVUS by OCT has been found to be higher in plaques with expansive
and may be superior for the identification of lipid pools (36). remodeling (40).
Several authors have evaluated the OCT appearance of coro-
nary plaques in different groups of patients, reporting higher Thickness of the Fibrous Cap
prevalence of lipid-rich plaques in ACS than in patients with Autopsy studies of sudden cardiac death victims have shown
stable angina (37) and no differences in the culprit plaque that the most frequent cause of the coronary occlusion is rup-
imaged by OCT between diabetic and nondiabetic patients ture of a thin-cap fibroatheroma (TCFA) plaque. Such lesions
[2/7/010/9:6:4] [370–387]
Figure 29.10 Bifurcation stent implantation in normal pigs. Serial, in vivo OCT (A) immediately after stenting and (B) at seven-day follow-up.
Tissue coverage is visible in various degrees from very thin coverage (C) to more pronounced, irregular coverage (D). Furthermore, a
persisted dissection flap can be observed at the proximal reference segment (E) (TD-OCT system, LightLab Imaging, Inc., Westford,
Massachusetts, U.S.). Abbreviations: OCT, optical coherence tomography; TD-OCT, time domain optical coherence tomography.
Table 29.3 Qualitative OCT Analysis of the Stent at Different Stent

Portions
Proximal Carina Distal
Dissection 2 (25.0) 1 (12.5) 1 (12.5)
Intraluminal tissue 2 (25.0) 0 0
Incomplete apposition 5 (62.5) 4 (50.0) 2 (25.0)
related to the thickness of the fibrous cap as measured by OCT

with yellow plaques often presenting thin caps (43,44).
Assessment of Collagen Composition

A fibrous cap is predominantly composed of collagen, synthe-
sized by intimal smooth muscle cells, which together impart
Figure 29.11 Bifurcation stent implantation in normal pigs (n ¼ 8).
mechanical integrity. Mechanisms that weaken the cap and
Quantitative OCT analysis of the stent area at different stent portion.
Abbreviation: OCT, optical coherence tomography.
potentially lead to plaque instability include collagen proteol-
ysis and impeded collagen synthesis, resulting in a net reduc-
tion in collagen content and thinning and disorganization of
collagen fiber orientation. Polarization-sensitive (PS) imaging
are characterized by a large necrotic core with a thin fibrous cap enhances OCT by measuring tissue birefringence, a property
usually <65 mm in thickness. Although conventional intracoro- that is elevated in biological tissues containing proteins with an
nary imaging techniques such as IVUS-VH do not have enough ordered structure, such as organized collagen (45–47). PS-OCT
resolution to evaluate in detail the fibrous cap, OCT has imaging can provide both, conventional grayscale OCT as well
demonstrated in correlation with histological examinations as PS-OCT images, in one single pullback through the coronary.
that it is able to provide accurate measurements of the thickness When light traverses birefringent tissue, light polarized along
of the fibrous cap (41,42). Therefore, it could be useful for the in directions parallel and perpendicular to the fiber orientation of
vivo detection of TCFA. In the study with IVUS, OCT, and the tissue travels at different velocities, incurring a relative
angioscopy in acute myocardial infarction patients by Kubo et phase retardation. The accumulated phase retardation is then
al., the incidence of TCFA was 83% and only OCT was able to displayed with respect to the tissue surface as a grayscale
estimate the fibrous cap thickness (mean 49 21 mm). Two image with black corresponding to 08 and white to 1808.
studies have reported that the plaque color by angioscopy is PS-OFDI birefringence has been demonstrated to be highly
[2/7/010/9:6:4] [370–387]
Figure 29.12 OCT image criteria for plaque type. Abbreviation: OCT, optical coherence tomography.
Figure 29.13 In vivo intravascular imaging of a calcified

coronary plaque. (A) Intravascular ultrasound shows an
eccentric, calcified plaque as echodense structure with
typical dorsal shadowing, while 1808 of vessel wall show
normal, three layer appearance. (B) Intravascular OCT
of the same plaque. By OCT, calcium appears signal
poor (dark) with sharp, well-delineated borders. (C) The
magnification reveals a small mural thrombus. Note,
that the calcium nodule is covered toward the lumen by
a thin layer and is not in direct contact to the lumen as
seen by IVUS. Abbreviations: OCT, optical coherence
tomography; IVUS, intravascular ultrasound.
related to total collagen content in atherosclerotic plaques as et al. evaluated the ability of OCT for the assessment of the
well as in fibrous plaques in vitro. It has been suggested that culprit lesion morphology in acute myocardial infarction in
the ability of OCT to measure changes in the fibrous cap comparison with IVUS and angioscopy. They found an inci-
thickness could be useful to assess the effect of statins in plaque dence of plaque rupture by OCT of 73%, significantly higher
stabilization (48,49). Furthermore, recent data suggest that than that detected by both angioscopy (47%, p ¼ 0.035) and
PS-OCT could be able to assess the collagen content and IVUS (40%, p ¼ 0.009). Intracoronary thrombus was observed in
smooth muscle cell density in the fibrous cap (45). This could all cases by OCT and angioscopy but was identified only in 33%
provide very valuable information about the mechanical sta- of patients by IVUS (49). Furthermore, Kume et al. demon-
bility of the fibrous cap, enabling the identification of lesions at strated that OCT might be able to distinguish between white
high risk of rupture. and red thrombus. Red thrombus appears in OCT as high-
backscattering structure with signal-free shadowing, while
Plaque Rupture and Intracoronary Thrombosis white thrombus appears as a low-backscattering structure
Plaque rupture with subsequent thrombosis is the most fre- (51). OCT could be helpful to identify the culprit lesion in
quent cause of ACS. OCT can identify intracoronary thrombus ACS and might provide additional information about the
and plaque rupture with high accuracy (50). Recently, Kubo underlying cause that lead to the plaque rupture (Fig. 29.14).
[2/7/010/9:6:4] [370–387]
Figure 29.14 OCT findings in a 74-year-old female

patient presenting with acute coronary syndrome,
Braunwald class IIIC with pos Trop T (0.57 mcg/L) and
CK (181 U/L) at the time of admission to the catheter-
ization laboratory. (A) Coronary angiography showed a
moderate narrowing of the mid left descending anterior
artery with TIMI III antegrade flow, the other epicardial
arteries were normal. Intracoronary OCT confirmed a
moderate lesion with approximately 50% area stenosis
(B) proximal reference lumen area 6.82 mm2, (C)
minimal lumen area 2.81 mm2, (D) distal lumen area
4.49 mm2. Detailed inspection of the OCT images
revealed (E) an eccentric plaque that showed a very
thin endothelial lining (<30 mm), (F) a site with plaque
rupture/ulceration, and (G) a fine fissure on the surface
of the plaque. Plaque erosion, rupture/ulceration, and
fissure have been associated with coronary thrombosis
in pathology series and could be the possible pathophy-
siological substrate for the clinical syndrome observed in
our patient (TD-OCT system, LightLab Imaging, Inc.,
Westford, Massachusetts, U.S.). Abbreviations: OCT,
optical coherence tomography; TD-OCT, time domain
optical coherence tomography.
Visualization of Macrophage Accumulation EQUIPMENT

Intense infiltration by macrophages of the fibrous cap is The equipment for intracoronary OCT generally consists an
another feature of the vulnerable plaques. An ex vivo study OCT imaging catheter, a motorized pullback device, and an
by Tearney et al. demonstrated OCT could be able to quantify imaging console, which contains the light source, signal proc-
macrophage within the fibrous cap (16). In vivo, it has been essing units, data storage, and display (54). The imaging cath-
demonstrated that unstable patients present a significantly eter is part of the sample arm of the interferometer described
higher macrophage density detected by OCT in the culprit above. The optical signal is transmitted by a single-mode fiber,
lesion than stable patients. Furthermore, in the same popula- which is fitted with an integrated lens microprism assembly to
tion, the sites of plaque rupture demonstrated a greater macro- focus the beam and direct it toward the tissue. The focus is
phage density than nonruptured sites (52). Raffel et al. reported approximately 1 mm outside the catheter. To scan the vessel
that macrophage density in the fibrous cap detected by OCT lengthwise, the catheter-imaging tip is pulled back while rotat-
correlated with the white blood cell count, and both para- ing, usually inside a transparent sheath, allowing to collect a 3D
meters could be useful to predict the presence of TCFA (53). dataset of the coronary artery. Both rotary and pullback motion
Figure 29.15 illustrates detailed OCT-based tissue characteriza- are driven proximally by a motor outside the patient. We will
tion over a long coronary segment including the distribution of describe the currently commercially available equipment for
macrophages, calcium, and lipid-rich tissue. time-domain OCT and the imaging procedure in detail.
[2/7/010/9:6:4] [370–387]
Figure 29.15 Longitudinal-cut view of an in vivo FD-OCT pullback

through a coronary artery. At the distal end, a coronary stent is
visible. Different components of the vessel wall are represented in a
color coded way. MGH OFDI system. Abbreviation: MGH OFDI, Figure 29.16 TD-OCT imaging (LightLab Imaging, Inc., Westford,
Massachusetts general Hospital optical frequency domain imaging. Massachusetts, U.S.)—the occlusive image acquisition technique.
Source: G. Tearney and B. Bouma, Wellman Center for Photo- The imaging catheter directs the infrared light into the tissue and
medicine, MGH, Boston, Massachusetts, U.S.A. returns the reflected light back to the optical engine. During imaging,
blood flow is limited by a dedicated, over the wire, low-pressure
(0.4 atm) occlusion balloon catheter (Helios, Goodman, Nagoya,
Japan), that is positioned proximally in the artery (A). The central
Equipment Time-Domain OCT lumen of the balloon occlusion catheter (B) allows for distal flush
The mobile M2/M2x OCT System cart (LightLab Imaging, Inc.) delivery during imaging of the target segment (C) with automated
contains the optical imaging engine and the computer. The pullback. The proximal end of the catheter has a chamber the
mouse, keyboard, two monitors, two storage drawers, and the captures the imaging core and allows it to move along the catheter
patient interface unit (PIU) are all mounted on top of the cart axis to perform an “automated pullback.” The outside of the imaging
(55) (Fig. 29.1). catheter is stationary with respect to the vessel wall. The imaging
core is rotated and translated inside of the external catheter sheath.
The ImageWireTM Abbreviations: TD-OCT, time-domain optical coherence tomography.
The imaging probe (ImageWireTM, LightLab Imaging, Inc.) has
a maximum outer diameter of 0.019 in. (with a standard
0.014 in. radiolucent coiled tip) and contains a single-mode
fiber optic core within a translucent sheath. The image wire is Occlusive Imaging Technique: Proximal Balloon Occlusion
connected at its proximal end to the imaging console that and Flush Delivery
permitted real-time data processing and 2D representation of The proximal occlusion balloon catheter (Helios, Goodman Co
the backscattered light in a cross-sectional plane. Since the Ltd., Nagoya, Japan) is an over-the-wire 4.4-Fr catheter (inner
imaging wire is not torquable, it can be advanced distal to diameter 0.025 in.), compatible with 6-Fr guiding catheters
the region of interest using the over-the-wire occlusion balloon (inner lumen diameter >0.071 in.), which is advanced distal
(for the occlusive technique, see next section) or a simple to region of interest using a conventional angioplasty guide-
microcatheter for the nonocclusive technique (e.g., Transit, wire (0.014 in.). The guidewire is then replaced by the OCT
Cordis, Johnson & Johnson, Miami, Florida, U.S.; or ProGreat, ImageWire (0.019 in. maximum diameter), and the occlusion
Terumo, Tokyo, Japan, with inner lumens > 0.020 in.; see sect. balloon catheter is withdrawn proximal to the segment to be
“Nonocclusive Imaging Technique”). Unlike an IVUS trans- assessed leaving the imaging wire in distal position. During
ducer, the optical sensor of the ImageWire is invisible under imaging acquisition, coronary blood flow is removed by con-
fluoroscopy and therefore one must estimate the correct posi- tinuous flush of Ringer’s lactate solution via the end hole of the
tion, using the distal 15-mm radiopaque tip of the ImageWire. occlusion balloon catheter at a flow rate of 0.5 to 0.7 mL/sec
When fully advanced, the sensor is located 6 to 7 mm proximal during simultaneous balloon inflation (0.5–0.7 atm). The vessel
to the radiopaque part and is easily confirmed by direct obser- occlusion time is limited to a maximum of 30 seconds to avoid
vation of the red light emitted when the wire is handled out of hemodynamic instability or arrhythmias. A 1.0 mm/sec pull-
the body. As there are no direct radiopaque markers for the back permits the assessment of an up to 30-mm long coronary
infrared sensor, it is possible to inadvertently miss imaging an segment with a frame rate of 15.4 frames/sec (56) (Fig. 29.16).
area of interest resulting in incomplete distal lesion edge
assessment. Imaging after stent implantation facilitates posi- Nonocclusive Imaging Technique
tioning because it is sufficient to advance the proximal end of With improvements in the acquisition speeds of OCT data
the radiopaque wire tip at least 1 cm distal to the stent struts to (currently in the vicinity of 3.0 mm/sec), blood can be evac-
image the entire stented segment. For imaging prior to treat- uated by continuous flush through the guiding catheter, thus
ment, it is important to note that the occlusion balloon is too doing away with the cumbersome proximal balloon occlusion
bulky to cross severe stenoses before predilatation and that the and thereby simplifying the acquisition process (57). Here, the
imaging wire should be advanced distal to the lesion to ensure OCT ImageWire is advanced carefully distal to the region of
that the segment of interest is fully visualized. In such circum- interest. As the fragile wire does not have the properties of a
stances, the nonocclusive technique is advantageous. guidewire, it needs to be directed distally using a single-lumen
[2/7/010/9:6:4] [370–387]
microcatheter (e.g., Transit, Cordis, Johnson & Johnson; strut and the vessel wall. Regardless of the pathophysiological
ProGreat, Terumo). With the wire in position, viscous iso- mechanism, the major concern in stent malapposition remains
osmolar contrast (Iodixanol 370, VisipaqueTM, GE Health in the assumption that areas of strut malapposition cause
Care, Ireland) at 378C is used to clear the artery from blood nonlaminar and turbulent blood flow characteristics, which in
and connected to the standard Y-piece of the guiding catheter. turn can trigger platelet activation and thrombosis. Here, pro-
The contrast is either injected manually through the Y-piece or spective, serial OCT observations immediately and at longer-
automatically using the Medrad pump. During continuous term follow-up after stenting may improve our understanding
contrast injection, the automated OCT pullback is performed of these complex mechanisms and shed light on the likely
at 3.0 mm/sec. The pullback is stopped after visualization of clinical significance of this phenomenon. The ongoing Optical
the region of interest or in case of significant signs of ischemia, Coherence Tomography Following Paclitaxel Eluting Stent
arrhythmia, or patient intolerance. Cross-sectional images are Implantation in Multivessel Coronary Artery Disease
acquired at 20 frames/sec. (OCTAXUS) will assess the proportion of uncovered and/or
After completion of the OCT study, the image wire is malapposed Taxus Libertè struts at different time points after
removed, and intracoronary nitrates are usually administered the implant, as measured by OCT.
according to local standards and an angiogram should be
taken. The procedure should be terminated if there is any
hemodynamic compromise during the infusion and OCT Stent Strut Tissue Coverage and Thrombosis
acquisition. Late stent thrombosis is poorly understood. It appears that this
condition is multifactorial with premature discontinuation of
dual antiplatelet therapy, stent underexpansion, hypersensitiv-
Equipment Fourier-Domain OCT ity, and lack of endothelial tissue coverage, all being impli-
The Fourier-domain OCT systems are under development and cated. The results of small observational OCT studies as
not yet commercially available. In the new generation of OCT
described earlier are compatible with evidence from animal
systems, the optical probe is integrated in a short monorail and human postmortem series, showing that DES cause impair-
catheter that can be advanced in the coronary artery over any ment in arterial healing, some with suggested incomplete
conventional 0.014-in. guidewire. The catheter profile varies re-endothelialization and persistence of fibrin(oid) possibly
from 2.4 Fr to 3.2 Fr and is compatible with 6-Fr guiding triggering late stent thrombosis (59,60). Pathological data in
catheters. The usable length is around 140 cm. The position human suggests that neointimal coverage of stent struts could
and number of the radiopaque markers vary for the different be used as a surrogate marker of endothelialization due to the
systems. During the pullback, the optic fiber probe is pulled good correlation between strut coverage and endothelializa-
along the catheter sheath. The dramatically faster pullback tion. However, OCT observations need to be interpreted with
speed allows imaging of long coronary segments with two to caution. OCT is limited by its resolution of 15 mm, which is
three seconds pullback during injection of flush (X-ray contrast greater than the thickness of an individual layer of endothelial
or diluted X-ray contrast) through the guiding catheter without cells. Therefore, coverage that is not visible by OCT does not
vessel occlusion. exclude the presence of an endothelial layer. Second, the pres-
ence of tissue coverage does not necessarily imply the presence
CLINICAL ASPECTS of a functionally intact endothelium. Early experimental stent
The interest in the long-term stent strut vessel wall interaction data showed that endothelial function can vary considerably
is manifold and includes the assessment of the stability of the and show evidence of damage when subjected to the Evan’s
acute result, the visualization of complex anatomy that is not blue dye exclusion test, even in the presence of a well-struc-
accessible by angiography or IVUS, and the clearer understand- tured neointimal layer (61).
ing of reasons for stent failures, when they do occur. However, OCT is the only imaging modality to date that
offers—within the discussed limits—the possibility to under-
stand tissue coverage and neointima formation in DES over
time (62). Clearly, larger stent trials with OCT at different time
DES Failure
periods are needed to obtain a representative assessment of the
Reasons for DES failure are poorly understood. With the reduc-
true time course of endothelial stent coverage of these stents.
tion of in-stent hyperplasia, other mechanisms of restenosis due
Recent improvements in OCT technology, with frequency-
to mechanical stent failure have become apparent. Of the two
domain OCT, will allow for a simple imaging procedure and
established first-generation DES, the SES (Cordis, Johnson &
offer the potential for large-scale, prospective studies, indis-
Johnson) has been particularly linked to cases of stent fracture,
pensable to address vexing clinical questions such as the rela-
likely as a result of its closed-cell design compared with other
tionship of DES deployment, vascular healing, the true time
DES employing an open-cell system (58). The higher imaging
course of endothelial stent coverage, and late stent thrombosis.
resolution of OCT compared with IVUS permits a detailed
This may also better guide the optimal duration of dual
assessment in such cases, as demonstrated recently by Shite
antiplatelet therapy that currently remains unclear and rather
et al.
empiric. The ongoing Optical Coherence Tomography Drug
Eluting Stent Investigation (OCTDESI) plans to evaluate the
Incomplete Strut Apposition completeness of strut coverage and vessel wall response to the
Stent strut malapposition remains another important consider- new generation JACTAX DES versus Taxus stent in de novo
ation. Postulated causes for stent strut malapposition are var- coronary artery lesions at six months post-index procedure.
ious and include incomplete stent expansion, stent recoil or The completeness of the coverage as well as the number of
fracture, late outward vessel remodeling, or the dissolution of uncovered stent struts per section, high resolution (*10–15 mm
thrombus that was compressed during PCI between the stent axial) will be assessed with intracoronary OCT. The Optical
[2/7/010/9:6:4] [370–387]
Coherence Tomography for Drug Eluting Stent Safety procedural time and in the incidence of chest pain and ECG
(ODESSA) study on the other hand will investigate the number changes during image acquisition. These side effects are
of uncovered and/or malapposed stent struts at overlapping expected to be further reduced by the introduction of Fourier-
versus nonoverlapping sites in DES versus BMS. domain OCT. In Fourier-domain OCT, high pullback speeds of
up to 40 mm/sec allow data acquisition of a long coronary
segment within few seconds and thus without introducing
DES Restenosis
relevant ischemia.
OCT can be very useful in the evaluation of the causes that
contribute to restenosis after DES implantation such as incom-
plete coverage lesion or gaps between stents. LIMITATIONS
Stent fracture (with subsequent defect of local drug Generally, the main limitations of intracoronary OCT compared
delivery) has also been related to restenosis in DES and could with the established IVUS technology consist first in the fact
be visualized with OCT (63). Nonuniform distribution of stent that light cannot penetrate blood. Thus, OCT requires clearing
struts can affect the drug concentration within the arterial wall of the artery from blood as additional step during the imaging
and therefore have an influence in restenosis in DES (64). This procedure. Secondly, the high resolution of OCT is at the
has been confirmed in preclinical and IVUS studies. The max- expense of penetration depth. More specific issues can impede
imum interstrut angle has been identified in IVUS as predictor intracoronary OCT quality and interpretation in clinical intra-
of intimal hyperplasia cross-sectional area. OCT allows the coronary OCT application. The most important practical limi-
assessment of strut distribution in vivo with high accuracy. A tations originating from coronary anatomy and artifacts are
study with phantom models showed how the strut distribution summarized below (56).
of SES and paclitaxel-eluting stents (PES) assessed by OCT
were significantly different, suggesting that SES maintained a
more regular strut distribution despite expansion (65). Vessel Tortuosity
Another field of clinical use for OCT might be the assess- Within the human body, peripheral arteries (the vascular access
ment of the performance of DES in complex coronary inter- sites) as well as the coronary arteries (imaging target) represent
ventions such as bifurcations. Buellesfeld et al. (66) reported the more or less tortuous structures. This requires high flexibility
nine-month OCT follow-up in a case of crush stenting with and steerability from the imaging device. This is not trivial
PES. Crush stenting results in three layers of metal in the given the fact that light transmission requires easily breakable
segment of the main vessel proximal to the stented side branch. fiber optics. Another consequence is on the image geometry. In
There has been concern that this could release a higher dose of the majority of cases, the imaging device will not be in a coaxial
drug locally with the potential adverse effect of delayed endo- and centered position within the target artery, which may affect
thelialization. In this report, OCT imaging showed the overpenetration depth, brightness, and resolution of the imaged
lapping struts layers in the crushed segment completely structure (Fig. 29.17).
covered by tissue. Furthermore, OCT allowed clear visualiza-
tion of the struts located in the ostium demonstrating a nonuni-
form distribution and different patterns of tissue coverage. The
effect of overlapping stents eluting different drugs in the devel-
opment of endothelial coverage might also be studied with
OCT.
OCT Safety
The applied energies in intravascular OCT are relatively low
(output power in the range of 5.0–8.0 mW) and are not consid-
ered to cause functional or structural damage to the tissue.
Safety issues thus seem mainly dependent on the need of blood
displacement for image acquisition. One recently published
study evaluated the safety and feasibility of OCT in 76 patients
in the clinical setting using the occlusive technique. Vessel
occlusion time was 48.3 13.5 seconds. The most frequent
complication was the presence of transient events, such as chest
discomfort, brady- or tachycardia, and ST-T changes on elec-
trocardiogram, all of which resolved immediately after the
procedure. There were no major complications, including myo-
cardial infarction, emergency revascularization, or death. The
authors reported that acute procedural complications such as
acute vessel occlusion, dissection, thrombus formation, embo- Figure 29.17 In vivo intracoronary OCT of an artery with concen-
lism, or vasospasm along the procedure-related artery were not tric intimal thickening. The imaging catheter is in a noncoaxial,
observed (67). Comparison of the occlusive versus the non- noncentered position. In consequence, the vessel wall segment
occlusive imaging method in a small patient cohort (n ¼ 40) did close to the OCT catheter in 6 o’clock position appears brighter
not show major complications and confirmed superiority of the and shows a finer structure as the opposite vessel wall segment in
nonocclusive method in ostial lesion assessment (68). In our 12 o’clock position. Abbreviation: OCT, optical coherence tomog-
experience, the introduction of the nonocclusive technique raphy.
in clinical practice has led to an important reduction in the
[2/7/010/9:6:4] [370–387]
Figure 29.18 In vivo intracoronary OCT of an artery with eccentric

atherosclerotic plaque. A normal vessel wall sector with typical
three-layer appearance is visible in 6 to 9 o’clock position, while
the remaining vessel circumference shows thick plaque formation.
Despite a centered position of the OCT imaging probe within the
lumen, the eccentric plaque cannot be completely penetrated and Figure 29.19 In vivo intracoronary OCT (A) illustrates the motion
thus, the adventitial layer is not visible from 12 to 5 o’clock position. artefact during heart cycle in the cross-sectional view, causing a
Abbreviation: OCT, optical coherence tomography. “seamline” indicated by the white arrow; (B) illustrates the motion
artefact caused in the longitudinal view. Note the irregular lumen
contour caused by 3D motion of the artery relative to the OCT
imaging catheter. Abbreviation: OCT, optical coherence tomography.
Coronary Caliber defined borders with diffuse transition to the surrounding

Coronary arteries represent relatively small structures for in tissue.
vivo imaging; they are, however, relatively big structures
compared with experimental OCT applications that often
focus on much smaller sample volumes. Epicardial arteries Motion During Heart Cycle
have a maximal lumen diameter of approximately 4 to 5 mm Epicardial arteries experience significant 3D motion during
in their proximal portion, and taper distally. Typically, arteries heart cycle. This affects (i) the vascular dimensions (with a
with a lumen diameter down to approximately 1.0 mm are variability of lumen area of 8% between systole and diastole
considered clinically relevant and accessible to standard imag- (71)), (ii) the OCT device position within the artery (transversal
ing equipment, such as coronary angiography and IVUS. Ide- and longitudinal motion), and (iii) the image acquisition time
ally, the penetration depth of intravascular OCT should be able (Fig. 29.19).
to cover the complete caliber range (Fig. 29.18). In coronary stents, typical stent imaging artifacts can be
observed (Fig. 29.20).
Plaque Geometry
Atherosclerotic coronary arteries contain a highly variable Shadowing Behind Stent Struts
degree of plaque deposition within the artery wall. Athero- The light source used for OCT is unable to penetrate metal
sclerotic plaque can form a concentric ring encroaching the resulting in dorsal shadowing behind the stent strut. When
lumen, but will be eccentric with a normal vessel wall sector or interpreting OCT images, the thickness of the whole stent strut
with relatively big differences in vessel wall thickness in the (including metal and polymer) must be taken into consider-
majority of cases (69). Intravascular OCT is hampered to pen- ation rather than only the visible endoluminal strut surface.
etrate advanced, thick plaque, irrespective of the position of the Shadowing also limits the interpretation of structures behind
OCT imaging device within the lumen. the stent strut, and this remains a limitation of OCT, particu-
larly also given its poor tissue penetration (<1.5 mm). Further-
more, the OCT imaging plane rarely intersects the stent struts
Plaque Composition perpendicularly, thereby resulting in shadows much larger
The limited penetration depth into the vessel wall can reduce than the actual width of the stent strut.
the sensitivity of OCT for different plaque components. An in
vitro study comparing OCT to histopathology reported mis-
classification in 41% of lesions predominantly due to a combi- Bright Reflections Saturating
nation of incomplete penetration depth into the vessel wall and an Entire Line (Spikes)
a resulting difficulty to distinguish calcium deposits from lipid If the imaging beam hits a strut perpendicularly, it reflects a
pools (70). Calcium deposits as well as lipidic tissues appear very large fraction of the beam back toward the catheter. This
signal-poor by OCT. These two tissue types can be discrimi- strong signal may saturate the detector registering the interfero-
nated by the tissue borders, calcium typically shows very gram, producing a readily recognizable artifact of bright radial
sharp, well-delineated borders, whereas lipid shows poorly streaks centered on struts.
[2/7/010/9:6:4] [370–387]
Figure 29.20 OCT artefacts in stent imaging. (A) Circumferential stent struts are shown covered by a thin layer of tissue. Dorsal shadowing
is evident behind the stent struts (s) with only the luminal surface of the strut visible with OCT. (B) Circumferential stent struts with shadowing
and bright reflections caused by saturation of the detector registering the interferogram (arrow). (C) Arrow indicates a double reflection
between stent and catheter. This can occur if a stent strut is imaged face-on, and reflects the OCT beam specularly. Shadows of other stent
struts are marked by s; stent struts that do not show up as the typical bright dot, but do cast a shadow, are imaged under an oblique angle. * is
a guidewire artefact. White bar is 1 mm. Abbreviation: OCT, optical coherence tomography.
Multiple Reflections (Stent-Catheter-Stent) acquisition in vivo. Similarly, interobserver variability in com-

In a similar geometry, near specular reflection, light may plex vessel anatomy as represented by chronic coronary stents
bounce back between catheter and strut more than once. The was very low: the absolute and relative difference between
optical catheter itself reflects part of the received light back into lumen area measurements derived from two observers was low
the tissue. Strong reflectors, such as struts, may produce an (0.02 0.10 mm2; 0.3 0.5%, respectively) with excellent
appreciable signal in the second reflection. The optical path correlation confirmed by linear regression analysis (R2 0.99; p
length of light in the secondary reflection is double that of the < 0.001). Similarly, in vivo measurements demonstrated a high
primary feature. Hence, a double reflection will show up as an correlation with the main source of interobserver variation
apparent second strut appearing behind the first at twice the occurring as a result of coronary dissection and motion artifact.
distance from the catheter. The absolute and relative difference between measurements
were 0.11 0.33 mm2 (1.57 0.05%) for lumen area (R2 0.98; p
< 0.001), 0.17 0.68 mm2 (1.44 0.08%) for stent area (R2 0.94;
SPECIAL ISSUES/CONSIDERATIONS/ p < 0.001), and 0.26 0.72 mm2 (14.08 0.37%) for neointimal
CONTRADICTIONS area (R2 0.78; p < 0.001).
The unique high-resolution OCT imaging modality permits the
analysis of coronary structures in great detail. The sharp con-
trast between the lumen during flushing and the vessel wall CONCLUSIONS
allows for a relatively easy image interpretation. It is notewor- OCT has caused intense interest in interventional cardiology.
thy that indeed OCT has a high accuracy when compared with Its application to the assessment of coronary stents has been
histomorphometry as discussed earlier, but it also has a greeted with strong enthusiasm and is now also being incor-
remarkably high reproducibility in the clinical setting. porated into large multicentre randomized stent trials aimed at
The measurement accuracy of intracoronary OCT has complementing angiographic and clinical endpoints. Such
been established in postmortem human coronary arteries and applications are currently unique to OCT and, despite the
showed good correlation to histomorphometry (72). However, recent progress of noninvasive techniques such as 64 multislice
compared to ex vivo imaging, quantitative analysis of in vivo computed tomography (MSCT), conventional stents still repre-
intracoronary imaging is more complicated due to the presence sent a challenge to distinguish lumen and intimal hyperplasia
of blood and motion artifacts during cardiac cycle. Further- within the stent. Reports showing that reconstruction of MSCT
more, the OCT dataset acquired during motorized pullback in images using specific kernels offer good correlation with
vivo is much larger than local imaging of selected cross- angiography are encouraging but unlikely to make this tech-
sections as performed in postmortem studies. A pullback nique a reliable alternative at the present time.
through the region of interest is necessary to visualize the 3D The ability to provide high-resolution imaging in vivo is
morphology of the coronary artery. We recently reported on the most significant concept circumventing the limitations of
standardized automated quantification processes for intracoro- other imaging modalities such as IVUS or the need for multiple
nary OCT pullback data (20). The interobserver variability for animal studies. Refinements in acquisition speeds with OFDI
lumen dimensions as measured by computer-assisted quanti- will also make the technique less procedurally demanding and
tative optical coherence tomography was extremely low and in thus able to be applied to many more centers, thereby remain-
a similar range for both in vitro as well for in vivo studies, ing a key tool in the armamentarium of researchers and
despite the occurrence of motion-induced artifacts during the interventional cardiologists alike.
[2/7/010/9:6:4] [370–387]
REFERENCES 24. Chen BX, Ma FY, Luo W, et al. Neointimal coverage of bare-metal
1. Huang D, Swanson E, Lin C, et al. Optical coherence tomography. and sirolimus-eluting stents evaluated with optical coherence
Science 1991; 254:1178–1181. tomography. Heart 2008; 94(5):566–570.
2. Brezinski ME, Tearney GJ, Bouma BE, et al. Imaging of coronary 25. Zimarino M, Prati F, Stabile E, et al. Optical coherence tomogra-
artery microstructure (in vitro) with optical coherence tomography. phy accurately identifies intermediate atherosclerotic lesions—an
Am J Cardiol 1996; 77:92–93. in vivo evaluation in the rabbit carotid artery. Atherosclerosis
3. Michelson A, Morley E. On the relative motion of the earth and 2007; 193:94–101.
the luminiferous aether. Philos Mag 1887; S5:449–463. 26. Regar E, van Beusekom HM, van der Giessen WJ, et al. Images in
4. Schmitt JM, Knuttel A, Yadlowsky M, et al. Optical-coherence cardiovascular medicine. Optical coherence tomography findings
tomography of a dense tissue: statistics of attenuation and back- at 5-year follow-up after coronary stent implantation. Circulation
scattering. Phys Med Biol 1994; 39:1705–1720. 2005; 112:e345–e346.
5. Schmitt JM. Optical coherence tomography (OCT): a review. IEEE 27. Tanimoto S, Aoki J, Serruys PW, et al. Paclitaxel-eluting stent
J Select Topics Quantum Electron 1999; 5:1205–1215. restenosis shows three-layer appearance by optical coherence
6. Yun S, Tearney G, de Boer J, et al. High-speed optical frequency- tomography. EuroIntervention 2006; 1:484.
domain imaging. Opt Express 2003; 11:2953–2963. 28. van Beusekom HM, Saia F, Zindler JD, et al. Drug-eluting stents
7. Chinn SR, Swanson EA, Fujimoto JG. Optical coherence tomog- show delayed healing: paclitaxel more pronounced than siroli-
raphy using a frequency-tunable optical source. Opt Lett 1997; mus. Eur Heart J 2007; 28:974–979.
22:340–342. 29. Meer FJvd, Faber DJ, Cilesiz I, et al. Temperature-dependent
8. Rollins AM, Kulkarni MD, Yazdanfar S, et al. In vivo video rate optical properties of individual vascular wall components mea-
optical coherence tomography. Opt Express 1998; 3:219–229. sured by optical coherence tomography. J Biomed Opt 2006;
9. Leitgeb R, Hitzenberger CK, Fercher AF. Performance of fourier 11:041120.
domain vs. time domain optical coherence tomography. Opt 30. Ormiston JA, Serruys PW, Regar E, et al. First–in-man evaluation
Express 2003; 11:889–894. of a bioabsorbable-everolimus eluting coronary stent system (BVS)
10. Choma M, Sarunic M, Yang C, et al. Sensitivity advantage of in the treatment of patients with single de-novo native coronary
swept source and Fourier domain optical coherence tomography. artery lesions: the ABSORB trial. Lancet 2008; 371:899–907.
Opt Express 2003; 11:2183–2189. 31. Yabushita H, Bouma BE, Houser SL, et al. Characterization of
11. de Boer JF, Cense B, Park BH, et al. Improved signal-to-noise ratio human atherosclerosis by optical coherence tomography. Circula-
in spectral-domain compared with time-domain optical coherence tion 2002; 106:1640–1645.
tomography. Opt Lett 2003; 28:2067–2069. 32. Rieber J, Meissner O, Babaryka G, et al. Diagnostic accuracy of
12. Yun SH, Tearney GJ, Vakoc BJ, et al. Comprehensive volumetric optical coherence tomography and intravascular ultrasound for
optical microscopy in vivo. Nat Med 2006; 12:1429–1433. the detection and characterization of atherosclerotic plaque com-
13. Lim H, Mujat M, Kerbage C, et al. High-speed imaging of human position in ex-vivo coronary specimens: a comparison with his-
retina in vivo with swept-source optical coherence tomography. tology. Coron Artery Dis 2006; 17:425–430.
Opt Express 2006; 14:12902–12908. 33. Patwari P, Weissman NJ, Boppart SA, et al. Assessment of coro-
14. Hong MK, Mintz GS, Lee CW, et al. Late stent malapposition after nary plaque with optical coherence tomography and high-fre-
drug-eluting stent implantation: an intravascular ultrasound anal- quency ultrasound. Am J Cardiol 2000; 85:641–644.
ysis with long-term follow-up. Circulation 2006; 113:414–419. 34. Kawasaki M, Bouma BE, Bressner J, et al. Diagnostic accuracy of
15. Tanabe K, Serruys PW, Degertekin M, et al. Incomplete stent optical coherence tomography and integrated backscatter intra-
apposition after implantation of paclitaxel-eluting stents or bare vascular ultrasound images for tissue characterization of human
metal stents: insights from the randomized TAXUS II trial. Circu- coronary plaques. J Am Coll Cardiol 2006; 48:81–88.
lation 2005; 111:900–905. 35. Kume T, Akasaka T, Kawamoto T, et al. Assessment of coronary
16. Tearney GJ, Yabushita H, Houser SL, et al. Quantification of arterial plaque by optical coherence tomography. Am J Cardiol
macrophage content in atherosclerotic plaques by optical coher- 2006; 97:1172–1175.
ence tomography. Circulation 2003; 107(1):113–119. 36. Jang IK, Bouma BE, Kang DH, et al. Visualization of coronary
17. Tanigawa J, Barlis P, Dimopoulos K, et al. Optical coherence atherosclerotic plaques in patients using optical coherence tomog-
tomography to assess malapposition in overlapping drug-eluting raphy: comparison with intravascular ultrasound. J Am Coll
stents. EuroIntervention 2008; 3:580–583. Cardiol 2002; 39:604–609.
18. Tanigawa J, Barlis P, Kaplan S, et al. Stent strut apposition in 37. Jang IK, Tearney GJ, MacNeill B, et al. In vivo characterization of
complex lesions using optical coherence tomography. Am J Cardiol coronary atherosclerotic plaque by use of optical coherence
2006; 98(suppl 1):97M. tomography. Circulation 2005; 111:1551–1555.
19. Regar E, Schaar J, Serruys PW. Images in cardiology. Acute recoil 38. Chia S, Raffel OC, Takano M, et al. Comparison of coronary
in sirolimus eluting stent: real time, in vivo assessment with plaque characteristics between diabetic and non-diabetic subjects:
optical coherence tomography. Heart 2006; 92:123. An in vivo optical coherence tomography study. Diabetes Res Clin
20. Tanimoto S, Rodriguez-Granillo G, Barlis P, et al. A novel Pract 2008; 81:155–160.
approach for quantitative analysis of intracoronary optical coher- 39. Chia S, Christopher Raffel O, Takano M, et al. In-vivo comparison
ence tomography: high inter-observer agreement with computer- of coronary plaque characteristics using optical coherence tomog-
assisted contour detection. Catheter Cardiovasc Interv 2008; raphy in women vs. men with acute coronary syndrome. Coron
72:228–235. Artery Dis 2007; 18:423–427.
21. Matsumoto D, Shite J, Shinke T, et al. Neointimal coverage of 40. Kume T, Okura H, Kawamoto T, et al. Relationship between
sirolimus-eluting stents at 6-month follow-up: evaluated by opti- coronary remodeling and plaque characterization in patients with-
cal coherence tomography. Eur Heart J 2007; 28:961–967. out clinical evidence of coronary artery disease. Atherosclerosis
22. Takano M, Inami S, Jang IK, et al. Evaluation by optical coherence 2008; 197:799–805.
tomography of neointimal coverage of sirolimus-eluting stent 41. Kume T, Akasaka T, Kawamoto T, et al. Measurement of the
three months after implantation. Am J Cardiol 2007; 99:1033–1038. thickness of the fibrous cap by optical coherence tomography. Am
23. Takano M, Yamamoto M, Inami S, et al. Long-term follow-up Heart J 2006; 152:755 e1–e4.
evaluation after sirolimus-eluting stent implantation by optical 42. Cilingiroglu M, Oh JH, Sugunan B, et al. Detection of vulnerable
coherence tomography: do uncovered struts persist? J Am Coll plaque in a murine model of atherosclerosis with optical coher-
Cardiol 2008; 51:968–969. ence tomography. Catheter Cardiovasc Interv 2006; 67:915–923.
[2/7/010/9:6:4] [370–387]
43. Takano M, Jang IK, Inami S, et al. In vivo comparison of optical 58. Lee SH, Park JS, Shin DG, et al. Frequency of stent fracture as a
coherence tomography and angioscopy for the evaluation of cor- cause of coronary restenosis after sirolimus-eluting stent implan-
onary plaque characteristics. Am J Cardiol 2008; 101:471–476. tation. Am J Cardiol 2007; 100:627–630.
44. Kubo T, Imanishi T, Takarada S, et al. Implication of plaque color 59. Finn AV, Joner M, Nakazawa G, et al. Pathological correlates of
classification for assessing plaque vulnerability. JACC Cardiovasc late drug-eluting stent thrombosis: strut coverage as a marker of
Interv 2008; 1:74–80. endothelialization. Circulation 2007; 115:2435–441.
45. Nadkarni SK, Pierce MC, Park BH, et al. Measurement of collagen 60. Finn AV, Nakazawa G, Joner M, et al. Vascular responses to drug
and smooth muscle cell content in atherosclerotic plaques using eluting stents: importance of delayed healing. Arterioscler
polarization-sensitive optical coherence tomography. J Am Coll Thromb Vasc Biol 2007; 27:1500–1510.
Cardiol 2007; 49:1474–1481. 61. van Beusekom HM, Whelan DM, Hofma SH, et al. Long-term
46. Nadkarni SK, Bouma BE, de Boer J, et al. Evaluation of collagen in endothelial dysfunction is more pronounced after stenting than
atherosclerotic plaques: the use of two coherent laser-based imag- after balloon angioplasty in porcine coronary arteries. J Am Coll
ing methods. Lasers Med Sci 2009; 24(3):439–445. Cardiol 1998; 32:1109–1117.
47. Oh WY, Yun SH, Vakoc BJ, et al. High-speed polarization sensitive 62. Barlis P, DiMario C, van Beusekom HMM, et al. Novelties in
optical frequency domain imaging with frequency multiplexing. cardiac imaging—optical coherence tomography (OCT). EuroIn-
Opt Express 2008; 16:1096–1103. tervention 2008; 4:C22–C26.
48. Chia S, Raffel OC, Takano M, et al. Association of statin therapy 63. Shite J, Matsumoto D, Yokoyama M. Sirolimus-eluting stent frac-
with reduced coronary plaque rupture: an optical coherence ture with thrombus, visualization by optical coherence tomogra-
tomography study. Coron Artery Dis 2008; 19:237–242. phy. Eur Heart J 2006; 27:1389.
49. Kubo T, Imanishi T, Takarada S, et al. Assessment of culprit lesion 64. Suzuki Y, Ikeno F, Yeung AC. Drug-eluting stent strut distribu-
morphology in acute myocardial infarction: ability of optical tion: a comparison between Cypher and Taxus by optical coher-
coherence tomography compared with intravascular ultrasound ence tomography. J Invasive Cardiol 2006; 18:111–114.
and coronary angioscopy. J Am Coll Cardiol 2007; 50:933–939. 65. Hasegawa T, Ako J, Ikeno F, et al. Comparison of nonuniform
50. Fercher AF, Mengedoht K, Werner W. Eye-length measurement strut distribution between two drug-eluting stent platforms.
by interferometry with partially coherent light. Opt Lett 1988; J Invasive Cardiol 2007; 19:244–246.
13:186–188. 66. Buellesfeld L, Lim V, Gerckens U, et al. Comparative endoluminal
51. Kume T, Akasaka T, Kawamoto T, et al. Assessment of coronary visualization of TAXUS crush-stenting at 9 months follow-up by
arterial thrombus by optical coherence tomography. Am J Cardiol intravascular ultrasound and optical coherence tomography.
2006; 97:1713–1717. Z Kardiol 2005; 94:690–694.
52. MacNeill BD, Jang IK, Bouma BE, et al. Focal and multi-focal 67. Yamaguchi T, Terashima M, Akasaka T, et al. Safety and feasibil-
plaque macrophage distributions in patients with acute and stable ity of an intravascular optical coherence tomography image wire
presentations of coronary artery disease. J Am Coll Cardiol 2004; system in the clinical setting. Am J Cardiol 2008; 101:562–567.
44:972–979. 68. Kataiwa H, Tanaka A, Kitabata H, et al. Safety and usefulness of
53. Raffel OC, Tearney GJ, Gauthier DD, et al. Relationship between a non-occlusion image acquisition technique for optical coherence
systemic inflammatory marker, plaque inflammation, and plaque tomography. Circ J 2008; 72:1536–1537.
characteristics determined by intravascular optical coherence 69. Freudenberg H, Lichtlen PR. [The normal wall segment in coro-
tomography. Arterioscler Thromb Vasc Biol 2007; 27:1820–1827. nary stenoses—a postmortal study (author’s transl)]. Z Kardiol
54. Regar E, Schaar JA, Mont E, et al. Optical coherence tomography. 1981; 70:863–869.
Cardiovasc Radiat Med 2003; 4:198–204. 70. Manfrini O, Mont E, Leone O, et al. Sources of error and inter-
55. Regar E, Schaar J, van der Giessen W, et al. Real-time, in-vivo pretation of plaque morphology by optical coherence tomogra-
optical coherence tomography of human coronary arteries using a phy. Am J Cardiol 2006; 98:156–159.
dedicated imaging wire. Am J Cardiol 2002; 90(suppl 6A):129H. 71. Weissman NJ, Palacios IF, Weyman AE. Dynamic expansion of the
56. Regar E, van Leeuwen AMGJ, Serruys PW, eds. Optical Coherence coronary arteries: implications for intravascular ultrasound meas-
Tomography in Cardiovascular Research. London: Informa urements. Am Heart J 1995; 130:46–51.
Healthcare, 2007. 72. Kume T, Akasaka T, Kawamoto T, et al. Assessment of coronary
57. Prati F, Cera M, Ramazzotti V, et al. From bench to bedside: a intima—media thickness by optical coherence tomography: com-
novel technique of acquiring OCT images. Circ J 2008; 72:839–843. parison with intravascular ultrasound. Circ J 2005; 69:903–907.
[26/6/010/22:42:14] [388–403]
30
Percutaneous coronary intervention—general principles

Jack P. Chen, Dongming Hou, Lakshmana Pendyala, and Spencer B. King
INTRODUCTION LESION CLASSIFICATION

Percutaneous coronary intervention (PCI) was made possible The ACC/AHA lesion classification scheme has been used to
by the pioneering work of Andreas Gruentzig, who in 1977 predict the success rate and risk of balloon angioplasty for
performed the first coronary angioplasty. That first patient is specific anatomic subsets (Table 30.2) (8). It incorporates angio-
still doing well more than 30 years later. Over that time span, graphic characteristics including angulation (both proximal to
however, there have been dramatic improvements in the tech- and at the lesion site), location, existence and importance of
nology available to interventional cardiologists. In addition, side branches, length, calcification, and presence of chronic
advances in the management of thrombotic complications, total occlusion. Type B lesions are further subcategorized as
vascular recoil, acute closure, and restenosis have been dra- B1 or B2 if one or more than one B characteristic is present.
matic. Currently, more than two million PCI procedures are Major technical and device advances during the ensuing years
performed annually, and PCI has significantly eclipsed coro- have greatly improved the success rate and safety for PCI and
nary artery bypass graft surgery (CABG) as the leading revas- thus reduced the predictive accuracy of this classification sys-
cularization method for coronary artery disease (CAD). tem. In an analysis from the Society for Cardiovascular Angiog-
raphy and Interventions (SCAI) registry involving 41,071
single-vessel PCIs from 1993 to 1996, Krone and colleagues
determined that the single most predictive lesion feature was
INDICATIONS FOR PCI patency (9). Specifically, the success rate of a patent class B
In general, elective PCI may be indicated in patients with stable lesion was similar to that of any class A lesion; and the success
angina or ischemic stress tests, and suitable coronary anatomy. rate of a patent class C lesion surpassed that of an occluded
Primary PCI, when performed promptly, improves clinical class B lesion. Nonetheless, it is clearly recognized that no
outcomes compared with fibrinolytic therapy in patients with anatomic classification scheme can comprehensively assess
ST segment elevation myocardial infarction (STEMI). In non–ST the complexity of a specific lesion; indeed, the success and
elevation myocardial infarction (NSTEMI) and unstable angina, risk of PCI is also dependent on the patient’s clinical status, the
an invasive approach is recommended by the American Col- operator’s expertise, and pharmacologic and device selections.
lege of Cardiology/American Heart Association (ACC/AHA)
guidelines in patients with high-risk features (1).
The thrombolysis in myocardial infarction (TIMI) risk EQUIPMENT
score is a prognostic tool for acute coronary syndrome (ACS) Guiding Catheters
patients (Table 30.1) (2). Similarly, the Global Registry of Acute Most coronary guiding catheters range from 6 to 8 French in
Coronary Events (GRACE) risk score includes eight variables: diameter, although the former predominates in current prac-
advanced age, Killip class, systolic blood pressure, ST segment tice. While seemingly simplistic in design when compared with
deviation, cardiac arrest during presentation, serum creatinine other PCI equipment, the guiding catheter may be the most
level, elevated cardiac markers, and heart rate (3). A meta- important determinant of procedural success. Stable guiding
analysis reported an 18% relative risk reduction in the com- catheter support is crucial to ensure delivery of the interven-
bined end point of death or myocardial infarction (MI) in high- tional device to the lesion, especially in the presence of inhos-
risk non–ST elevation (NSTEMI) ACS patients initially treated pitable anatomy such as severe calcification or tortuosity.
with the invasive compared with the conservative strategy, The general principle in guiding catheter selection is to
despite a slight increase in early in-hospital mortality (4). Not use the smallest diameter and least aggressive design necessary
all studies have confirmed these findings, however. The Inva- for the specific coronary anatomy. Conservative, atraumatic tip
sive Versus Conservative Treatment in Unstable coronary conformations, such as Judkins right and left curves, are gen-
Syndromes (ICTUS) study (5) randomized 1200 NSTEMI erally selected for low complexity cases. Gentle, shallow ostial
patients to an initial invasive versus conservative strategy engagement minimizes potential coronary artery trauma and
and found no difference in the composite end point at one risk of catheter-induced dissection. When faced with more
year. Nonetheless, the majority of large randomized trials, challenging anatomy, enhanced back-up support is required.
including Fragmin and Fast Revascularization During Instabil- The moderate support designs are commonly known as the
ity in Coronary Artery Disease (FRISC) (6) and Treat Angina “extra back-up” group, while even deeper engagement can be
with Aggrastat and Determine Cost of Therapy with an Inva- achieved with the Amplatz catheters. Meticulous manipulation
sive or Conservative Strategy (TACTICS-TIMI 18) (7) support is required to avoid ostial left main or right coronary dissection,
the invasive approach for the high-risk patient cohort. which not infrequently leads to distal spiral dissection
[26/6/010/22:42:14] [388–403]
PERCUTANEOUS CORONARY INTERVENTION—GENERAL PRINCIPLES 389
Table 30.1 Thrombolysis in Myocardial Infarction Risk Factors lubricious coating also promotes wire entry into false channels
for Prognostication in Patients with Acute Coronary Syndrome (10). Short and long guidewires are designed for use with
l
rapid-exchange and over-the-wire devices, respectively.
Age 65 or greater
l Three or more coronary artery disease risk factors Varying tip stiffness allows for penetration into hard
l Known coronary stenosis of 50% or greater on lesion caps frequently associated with chronic total occlusions.
previous angiography Additionally, tapered tips may further improve penetration.
l ST segment deviation >0.05 mV on initial ECG Extreme caution must be exercised with these specialty wires,
l At least 2 anginal episodes in prior 24 hr as coronary dissection or perforation can result (10). Supportive
l Aspirin use in prior 7 days shaft wires are used to straighten tortuous arteries to allow
l Elevated cardiac markers passage of higher profile devices such as atherotomes or stents.
Not infrequently, however, “pseudolesions” may result from
arterial pleating and may be difficult to angiographically
discern from iatrogenic trauma. If intracoronary nitroglycerin
Table 30.2 American College of Cardiology/American Heart fails to resolve this finding, an over-the-wire catheter can be
Association Lesion Classifications exchanged for the stiff wire. This maneuver will resolve the
pleating artifact, while maintaining distal access in case further
Type A (high success rate >85%; low risk)
work is necessary.
l Discrete (<10 mm in length)
l Concentric
l Easily accessible Balloon Dilatation Catheters
l Relatively straight (<458) The original PCI device introduced by Gruentzig in 1977, the
l Smooth contour balloon catheter, remains the workhorse for the intervention-
l Minimal to no calcification alist. In the modern era, balloons are predominately used for
l Nonoccluded
l
pre- and postdilatation of stents.
Nonostial in location
l
Compliant balloons, made of soft materials such as
No major branch involvement
l No thrombus polyolefin copolymer which allow varying diametric expansion
without exerting excessive pressure, are useful for predilatation
Type B (moderate success rate ¼ 60–85%; moderate risk) to prepare the intended lesion for stent deployment. Occasion-
l Tubular (10–20 mm in length) ally, small caliber branches may be inappropriate for stent
l Eccentric implantation and can be adequately treated with “plain old
l Moderate tortuosity of proximal segment balloon angioplasty” (POBA). Noncompliant balloons, fash-
l Moderate lesion angulation (45–908) ioned from stiff materials such as polyethylene terephtalate,
l Irregular contour
l
maintain a near-constant diameter in the face of increasing
Moderate to heavy calcification
l
inflation pressures, making them ideal devices for stent post-
Ostial in location
l Bifurcation location, necessitating double wires dilatation to ensure full expansion. Additionally, noncompliant
l Thrombus present balloons can be used to expand hard, calcific or fibro-calcific
l Total occlusion (<3 mo) plaques. Semi-compliant balloons display hybrid features of
both their compliant and noncompliant counterparts (10).
Type C (low success rate <60%; high risk)
When inflated to the nominal pressure, the balloon achieves
l Diffuse (>20 mm in length) its prespecified diameter.
l Severe tortuosity of proximal segment The monorail or rapid-exchange shaft design features a
l Extreme lesion angulation (>908)
l
short wire lumen, with the wire exit opening near the distal tip.
Inability to protect major side branch
l
This system allows use of a short guidewire, often preferred by
Degenerated vein grafts, with friable lesions
l Chronic total occlusion (>3 mo) operators for simpler cases. Conversely, the wire lumen for the
over-the-wire balloon catheter runs the entire shaft length.
Source: From Ref. 8. While the required long guidewire may be more tedious, this
system permits catheter exchange as well as distal intracoro-
nary drug administration through the central lumen.
propagation. At the conclusion of the procedure, disengage-
ment of Amplatz catheters should be performed under fluoro-
scopic observation and may require a slight forward push with Perfusion Balloon
rotation, as direct withdrawal may actually result in deep, Although perfusion balloons were initially developed to reduce
forceful distal advancement. ischemia during prolonged balloon inflations during the pre-
stent era, they are now largely relegated to sealing of coronary
perforation. Proximal inflow and distal outflow lumens allow
Guidewires distal perfusion during protracted inflation of the perforation
Structurally, guidewires are divided into spring-tip and plastic site.
varieties. The former can either be constructed entirely from
stainless steel or feature a nitinol tip. The spring-tip wires offer
excellent steerability by enhanced torque transmission and are Thrombectomy Devices
particularly useful in complex lesions such as chronic total PCI of thrombotic and/or atherosclerotic lesions can result in
occlusions. Plastic wires are hydrophilically coated and offer plaque fragmentation and distal embolization. This can lead to
the advantage of reduced friction, facilitating advancement diminished or absent distal perfusion, known as “slow flow” or
through tortuous or calcified anatomies. However, this same “no reflow.” The perfusion adequacy of the distal vasculature
[26/6/010/22:42:14] [388–403]
and subtended myocardium are measured by the TIMI flow reported a 42% relative reduction in major adverse cardiac
and myocardial blush scores, respectively. Pretreatment with events (MACE) for the treatment arm; this benefit was mainly
certain aspiration devices may decrease the incidence of slow due to reductions in MI (8.6% vs. 14.7%, p ¼ 0.008) and no
and no reflow and may improve procedural outcome. Recent reflow (3% vs. 9%, p ¼ 0.02).
data suggest that manual thrombus aspiration may improve Filter devices preserve perfusion during deployment, so
survival in STEMI patients (11). their use is not limited by ischemic time. However, they do not
A mechanical thrombectomy catheter, the AngioJet (Pos- trap small debris and nonparticulate vasoconstrictive substan-
sis Medical Corporation, Minneapolis, Minnesota, U.S.), utilizes ces. The FilterWire EX distal filtration device (Boston Scientific,
an Archimedes screw mechanism to create a proximally Natick, Massachusetts, U.S.) is a guidewire that terminates in
directed high-pressure jet, which generates a vacuum rheolytic a loop net with 110-mm pores. The FilterWire EX Randomized
action. The captured debris is then removed through the sys- Evaluation (FIRE) trial (17) demonstrated noninferiority of this
tem. The catheter is advanced distal to the thrombus and device with the GuardWire during PCI in saphenous vein grafts.
withdrawn at a rate of 0.5 mm/sec. Not infrequently, there is
concomitant bradycardia (especially in the right coronary sys-
tem) and transient ST segment elevation. Prophylactic tempo- Atherectomy Devices
rary right ventricular pacemaker placement has been Atherectomy or atheroablative devices are applicable in specific
recommended; recently, intravenous aminophylline infusion anatomic subsets such as calcified or thrombotic lesions, either
has been shown to prevent heart rate disturbances (12,13). as stand-alone therapy or, more commonly, as pretreatment to
Manual thrombus aspiration can be accomplished with facilitate stent deployment. Although quite varied in mecha-
the Export (Medtronic Corporation, Minneapolis, Minnesota, nism and design, these tools may share a slightly higher risk of
U.S.) or Pronto (Vascular Solutions Incorporated, Minneapolis, coronary perforation compared with balloon angioplasty.
Minnesota, U.S.) catheters. These similar rapid-exchange devi- Additionally, while a few trials have reported modest rest-
ces feature a second aspiration lumen. Gentle manual aspira- enotic advantages, no atherectomy device has clearly demon-
tion with a syringe is performed by the assistant while the strated superior clinical outcomes compared with balloon
distal tip is advanced through the stenosis. The attraction of angioplasty (18).
these devices is ease of set-up and use, as well as lack of
associated bradyarrhythmias. However, their efficacy with
large thrombus burden or late patient presentation remains Rotational Atherectomy
unclear (14,15). The rotational atherectomy (Boston Scientific, Natick, Massa-
chusetts, U.S., Fig. 30.2) system consists of a rapidly rotating
oval-shaped burr encrusted with tiny diamond chips. The
Distal Protection Devices mechanistic principle is that of differential cutting, whereby
Distal protection devices, including occlusion balloons and hard plaque is selectively ablated while the deflected soft
filters, trap and retrieve atheroemboli distal to the lesion. The vascular wall is left unharmed. The burr is welded on a drive
GuardWire system (Medtronic Corporation, Minneapolis, Min- shaft which is advanced over a proprietary 0.09-in guidewire.
nesota, U.S., Fig. 30.1) incorporates a wire-based distal occlu- The recommended burr-to-artery ratio is 0.7. Through gentle
sion balloon in combination with the Export aspiration catheter. pecking motions of 0.5 mm or less, the operator manipulates
The Saphenous Vein Graft Angioplasty Free of Emboli the burr with the advancer console at a typical rotational speed
Randomized (SAFER) trial (16) randomized 801 patients with
saphenous vein graft disease to stent deployment with and
without concomitant GuardWire protection. The investigators
Figure 30.1 The GuardWire distal protection device, used in

conjunction with the Export aspiration catheter. Figure 30.2 Rotational atherectomy burr.
[26/6/010/22:42:14] [388–403]
of approximately 140,000 rpm. When encountering decelera- actual plaque excision, rather than dilatation, should reduce
tions of more than 5000 rpm, indicating extreme lesion resis- the risk of side branch compromise by plaque shift. This tech-
tance, operation should be paused. Further burring in this nique should be avoided in distal lesions, small branches,
scenario may result in vascular trauma from dissection, exces- degenerated SVGs, and calcified or tortuous vessels (13). Tech-
sive heat buildup, or embolization (18). nical difficulties and the need for large caliber guiding catheters
“Wire bias” refers to the phenomenon whereby greater have limited the widespread use of directional atherectomy.
cutting force is exerted on the inner curvature of a tortuous
vessel. While this occurrence may be advantageous for eccen-
tric lesions in this location, it can lead to excessive coronary
Laser Atherectomy
The XeCl Excimer utilizes ultraviolet energy with a wavelength
trauma. Significant bradycardia may result from microemboli-
of 308 nm to ablate debris through photochemical dissociation.
zation; hence, prophylactic temporary pacemaker placement is
The subsequent photoacoustic effect and heat generation elicits
recommended, especially for right coronary interventions (13).
formation of a vapor bubble, which ultimately implodes. Typ-
The Study to Determine Rotablator and Transluminal
ically, the repetition rate is set at 40 Hz and energy or fluence
Angioplasty Strategy (STRATAS) (19) evaluated aggressive
set up to 60 mJ/mm2. Initial issues with arterial dissection were
versus conservative rotational atherectomy strategies using
largely attributed to excessive heat generation in the blood- or
burr-to-artery ratios of 0.7 to 0.9 versus 0.7. While overall success
contrast-filled lumen. The revised technique of constant saline
rates were similar, the aggressive arm demonstrated significantly
infusion during activation has greatly reduced this complica-
higher restenosis (58% vs. 52%) and MI (11% vs. 7%) rates. In the
tion (25). The recommended catheter-to-artery ratio is 2:3.
Angioplasty Versus Rotational Atherectomy for Treatment of
Technique-wise, the catheter is advanced at a slow rate of 0.5
Diffuse In-Stent Restenosis Trial (ARTIST) (20) the investigators
to 1.0 mm/sec, with simultaneous gentle intracoronary saline
compared rotational atherectomy with balloon angioplasty for
infusion at 2 to 3 cc/sec.
treatment of in-stent restenosis (ISR) in 298 patients. Surprisingly,
The Excimer Laser Rotational Atherectomy Balloon
in-hospital complication and restenosis rates were higher with
Angioplasty Comparison (ERBAC) trial (26) randomized
rotational atherectomy, and the six-month event-free survival
685 patients to PCI with three therapeutic modalities. Slightly
was lower (79.6% vs. 91.1%). Nonetheless, rotational ablation of
higher overall MACE and restenosis rates were observed for
calcific, resistant lesions remains a valuable tool in some complex
laser subjects. However, the Amsterdam-Rotterdam (AMRO)
cases.
trial (27) found no significant differences in six-month MACE
between balloon and laser-treated patients. Modern usage of
laser atherectomy is aimed at thrombotic, friable saphenous
Directional Atherectomy
vein graft, bifurcation, and moderately calcific lesions. Even
Since its inception, versions of directional atherectomy devices
moderately tortuous vessels should be avoided, however, as
have progressed through several iterations. Presently, the Flexi-
wire bias can result in perforation from asymmetric arterial
cut catheter (Abbott Vascular Corporation, Abbott Park, Illi-
wall contact with the vapor bubble.
nois, U.S.) is approved for coronary use. Diameter sizes are 2.5
to 2.9 mm, 3.0 to 3.4 mm, and 3.5 to 3.9 mm (20). The excised
plaque debris is pushed into the distal nose-cone and subse- Cutting Balloon
quently retrieved when the compartment is filled. Selection of The cutting balloon catheter (Flextome, Boston Scientific Corpo-
guiding catheters is crucial, as a gentle curve is necessary to ration, Natick, Massachusetts, U.S.) incorporates longitudinal
allow smooth vessel entry. Under fluoroscopic visualization, stainless steel blades affixed to the balloon. When deployed,
the eccentrically located concave cutting window is directed discrete cuts made in the atheroma may avoid large uncon-
toward the plaque. A counterfacing compliant balloon is trolled dissections, resulting in less coronary trauma. Utilizing a
inflated to low pressures to maximize atherotome-to-plaque similar mechanistic principle, the AngioSculpt catheter (Angio-
contact. Early experience, however, revealed that some of the Score Corporation, Fremont, California, U.S.) features helical
immediate luminal gain was secondary to balloon stretch and wires surrounding the balloon, thus enhancing its flexibility to
failed to result in long-term benefit. access tortuous anatomy. Both devices have been used to treat
Despite initial enthusiasm, present-day utilization of ISR, bifurcation lesions, and moderately calcified plaques.
directional atherectomy has significantly declined. The Coro-
nary Angioplasty Versus Excisional Atherectomy Trial
(CAVEAT)-I (21) and Canadian Coronary Atherectomy Trial Bare-Metal Stents
(CCAT) (22) each reported higher immediate luminal gain after The advent of bare-metal stents (BMS) has dramatically
directional atherectomy compared with balloon angioplasty in reduced the significant target vessel revascularization (TVR)
native coronary lesions. However, no significant differences rates previously associated with balloon angioplasty. Up to this
were observed for follow-up restenosis rates. Similar findings point, no other nonballoon PCI device had clearly demon-
were seen in CAVEAT-II (23), which enrolled patients with strated a superior impact on restenosis. By scaffolding the
saphenous vein graft disease. Additionally, atherectomy- dilated segment, BMS eliminated the elastic recoil ubiquitously
treated patients in this study suffered a higher rate of NSTEMI. seen with balloons. Ironically, BMS implantation actually trig-
Even in the stent era, the Atherectomy Before Multi-link gers greater neointimal proliferation than balloon angioplasty.
Improves Lumen Gain and Clinical Outcomes (AMIGO) (24) However, the significantly larger initial luminal gain, coupled
trial failed to demonstrate either angiographic or clinical benefit with freedom from elastic recoil, results in a marked restenosis
of directional atherectomy prior to stenting versus stenting alone. and TVR advantage over balloon angioplasty.
Today, the main “niche” applications of this device Importantly, BMS have greatly enhanced the procedural
appear to be ostial lesions—especially left anterior descending safety of PCI. Dissections can often be successfully treated,
and left circumflex—and bifurcation lesions. In principle, dramatically decreasing the need for emergent CABG.
[26/6/010/22:42:14] [388–403]
Nonetheless, the implantation of an intracoronary prosthesis is mimics an erythrocyte outer layer, is associated with greater
not without risk. Nonendothelialized stent struts, as well as endothelialization than seen with either PES or SES. At
gaps between the stent and vascular wall, are potent niduses 24-month follow-up, Randomized Controlled Trial to Evaluate
for thrombosis. To ensure adequate stent expansion, meticulous the Safety and Efficacy of the Medtronic AVE ABT-578 Eluting
high-pressure postdilatation is often performed with a non- Drive Coronary Stent in De Novo NatiVe Coronary Artery
compliant balloon. A minimum of one month of dual antipla- Lesions (ENDEAVOR)-II trial observed that, in addition to
telet therapy is strongly recommended following BMS the expected superior TVR and MACE rates, the ZES arm
deployment (28). actually demonstrated a nonsignificantly lower stent thrombo-
sis rate of 0.5% versus 1.2% for BMS (37).
Drug-Eluting Stents Overall DES Safety and Efficacy

To overcome the enhanced endothelial hyperplasia induced by More recent data indicate that the late stent thrombosis risk
BMS implantation, drug-eluting stents (DES) were developed with DES is counterbalanced by a reduction in TVR, resulting
by incorporating antiproliferative coatings on stent struts. The in equivalent (if not slightly superior) MI, mortality, and overall
seminal Randomized Study with the Sirolimus-Coated Bx MACE rates compared with BMS. The Swedish Coronary
Velocity Balloon-Expandable Stent in the Treatment of Patients Angiography and Angioplasty Registry (SCAAR) (38) observed
with De Novo Native Coronary Artery Lesions (RAVEL) (29), equivalent long-term mortality risks for DES and BMS patients,
Sirolimus-Eluting Stent in De Novo Native Coronary Lesions despite a slight increase in late stent thrombosis. This report
(SIRIUS) (30), and Treatment of De Novo Coronary Disease was particularly noteworthy, given that initial findings from
Using a Single Paclitaxel-Eluting Stent (TAXUS) (31) trials this registry suggested a higher DES death rate, sparking an
reported superior clinical restenosis rates with sirolimus- early firestorm of anti-DES literature.
eluting stents (SES, Cordis, Miami Lakes, Florida, U.S.) and Recently, three separate analyses have further supported
paclitaxel-eluting stents (PES, Boston Scientific Corporation, the safety and efficacy of DES during long-term follow-up
Natick, Massachusetts, U.S.) compared with BMS Additonally, (39–41). Four-year data from over 5200 patients in the major
zotarolimus-eluting stents (ZES, Medtronic Corporation, randomized controlled SES and PES trials (RAVEL, SIRIUS,
Minneapolis, Minnesota, U.S.) and everolimus-eluting stents E-SIRIUS, C-SIRIUS, and TAXUS I-VI) demonstrated that death
(EES, Abbott Laboratories, Abbott Park, Illinois, U.S.) are now and MI rates were similar for both DES and BMS. Furthermore,
available for clinical use in the United States. despite a slight increase in late stent thrombosis with DES, the
Indeed, DES may challenge the long-term patency superi- overall stent thrombosis and MACE rates were similar for all
ority of CABG. The Arterial Revascularization Therapies Study stents. Most strikingly, the expected DES superiority in TVR
(ARTS)-II investigators compared SES therapy for multivessel reduction was reinforced; TVR rates for SES (7.8%) and PES
disease with historical data from patients assigned to the CABG (10.1%) were significantly lower than for BMS (23.6%,
arm in the preceding ARTS-I trial. The investigators found no p < 0.001) (41). Major contributors to stent thrombosis were
differences in MACE, mortality, TVR, or stroke at one year (32). male gender, continued tobacco use, overlapping stents, and
dual antiplatelet therapy noncompliance.
Concern About Late DES Thrombosis Another recent large meta-analysis documented persis-
While initial landmark trials demonstrated equivalent safety to tent and continued TVR benefits by 50% to 70% for up to four
BMS, subsequent follow-up suggested a late stent thrombosis risk. years, without notable differences in death or MI (42). Further-
Camenzind reported a significantly higher 6.3% combined death/ more, the incidences of ARC defined stent thrombosis were
MI rate for SES compared with 3.9% observed for BMS (33). similar (HR 1.00, 1.03, and 0.96 for SES vs. BMS, PES vs. BMS,
The Basel Stent Kosten Effiktivitats (Cost-Effectiveness) and SES vs. PES, respectively). Additionally, the use of SES was
Trial—Late Thrombotic Events (BASKET-LATE) study fol- associated with lower risk of MI, compared with PES (HR 0.83,
lowed 544 DES and 281 BMS patients for 18 months. DES p ¼ 0.045) or BMS (HR 0.81, p ¼ 0.030). Consistent with
patients experienced higher MACE from 6 to 18 months post previous findings, rates of stent thrombosis were higher for
implantation, including absolute excesses in late mortality BMS within one year and higher for DES beyond one year.
(1.2% vs. 0%, p ¼ 0.09) and nonfatal MI (4.1% vs. 1.3%, Notably, the numbers needed to treat to prevent one MACE
p ¼ 0.04) compared with BMS (34). event were 7 and 8 for SES and PES versus BMS, respectively.
To standardize definitions of stent thrombosis, the Aca- These benefits were similarly observed in diabetic patients.
demic Research Consortium (ARC) has classified stent throm- The safety and efficacy of DES in diabetic patients were
bosis according to the relative certainty of the diagnosis. further demonstrated in a meta-analysis of nine randomized
Definite thrombosis includes angiographic target vessel occlu- DES trials involving 1141 diabetic patients (43). DES was asso-
sion or intrastent thrombus. Probable thrombosis is defined as ciated with lower ISR (8% vs. 41%, OR 0.13, 95% CI, 0.09–0.20,
acute MI in the target vessel territory, while possible thrombo- p < 0.000,01) and TVR (8% vs. 27%, OR 0.23, 95% CI, 0.16–0.33,
sis represents any unexplained death in a patient with a p < 0.000,01). Interestingly, while overall stent thrombosis or
previously implanted DES (35). death rates were similar between groups, subsequent MI was
A four-year combined study from Switzerland and the more frequent in BMS patients (7.2% vs. 3.5%, p ¼ 0.02). The DES
Netherlands evaluated risk factors for stent thrombosis in over advantages prevailed regardless of insulin-dependency status.
8000 patients receiving SES or PES. The investigators reported It is imperative that all DES patients at low risk of
that diabetes was an independent risk factor for early stent bleeding receive one year of dual antiplatelet therapy (44,45).
thrombosis, while PES implantation, younger age, and ACS Additionally, prior to stent implantation, patients should be
were associated with late stent thrombosis (36). assessed for financial or clinical challenges (such as upcoming
ZES, with a more rheologically compatible design, rep- surgery) to prolonged dual antiplatelet therapy. Health care
resents the second generation of DES. Its polymer, which providers performing invasive procedures should be well
[26/6/010/22:42:14] [388–403]
Table 30.3 Recommendations for DAT in Patients with DES

1. DES should only be implanted in patients able to comply with 12 mo of DAT, after economic and clinical considerations.
2. For patients undergoing percutaneous coronary intervention who are likely to undergo invasive or surgical procedures within the next
12 mo, BMS implantation or stand-alone balloon angioplasty should be considered.
3. Health care providers should more thoroughly educate patients regarding the indications of DAT as well as the risks of premature
termination.
4. Patients should be advised to contact their cardiologist prior to interruption of DAT, even if instructed to do so by another health
care provider.
5. Health care providers who perform invasive or surgical procedures should be aware of the potentially serious sequelae of premature
DAT interruption. They should contact the patient’s cardiologist if unsure about the safety of therapy cessation.
6. Elective invasive or surgical procedures associated with significant bleeding risk should be postponed, if possible, until appropriate DAT
course is completed (12 mo for DES and 1 mo for BMS).
7. For DES patients undergoing invasive or surgical procedures mandating thienopyridine discontinuation, aspirin should be continued if at
all feasible. Thienopyridine should be reinitiated as soon as possible after the procedure.
8. Health care industry, insurers, U.S. Congress, and pharmaceutical industry should ensure that cost issues do not cause premature
thienopyridine discontinuation, as this can result in devastating consequences.
Abbreviations: DES, drug-eluting stents; BMS, bare-metal stents; DAT, dual antiplatelet therapy.
Table 30.4 Summary of DES Safety and Recommendations for DES Use
l When compared with BMS, DES are associated with a small but definite increase in risk of late stent thrombosis, predominately
manifested 1 yr after implantation.
l When used according to “on-label” indications, DES result in no overall increase in mortality or MI rates compared with BMS. This finding
is likely related to the marked decrease in repeat revascularization associated with DES.
l DES and BMS have the similar total mortality rates.
l Larger and longer premarket clinical trials are needed, with longer follow-up durations, more uniform stent thrombosis definitions, and
closer attention to DAT therapy.
l At this time, when deployed according to approved indications, the risks of DES do not outweigh their benefits compared with BMS.
l When used in nonapproved patient and anatomic subsets, a higher risk of DES thrombosis, MI, or death may be expected than observed
in previous trials.
l Because of limited data for “off-label” DES use (accounting for at least 60% of present-day DES implantation), more studies are needed
to determine the safety and efficacy of these devices. DES labeling should state that “off-label” use may not yield the same results as
those observed in the clinical trials which led to market approval.
l Studies suggest that prolonged DAT beyond product label recommendations may be beneficial.
l Optimal DAT duration is unknown, and continued DAT does not guarantee freedom from DES thrombosis.
l Patients at low risk of bleeding should continue DAT for 12 mo.
Abbreviations: DES, drug-eluting stents; BMS, bare-metal stents; MI, myocardial infarction; DAT, dual antiplatelet therapy.
versed in indications for dual antiplatelet therapy, and any approved for emergent bail-out treatment of coronary perfora-
plans for premature interruption should be discussed with the tion. Given the device’s high profile and relative rigidity, ade-
patient’s cardiologist. These recommendations are summarized quate catheter back-up is crucial. As the need for this stent is
in Table 30.3. rare, but invariably emergent, interventionalists should famil-
An expert panel concluded that DES use is associated with iarize themselves with its deployment technique. Trials involv-
a real, albeit rare, increased incidence of late stent thrombosis ing elective covered stent implantation have uniformly failed to
that frequently results in MI or death. However, when these demonstrate a restenosis advantage (46,47).
devices are used for “on-label” indications, there is no evidence
of increased stent thrombosis, MI, or death rates. Furthermore,
uninterrupted dual antiplatelet therapy for at least one year is
Intravascular Ultrasound
Intravascular ultrasound (IVUS) interrogation allows for intra-
strongly recommended in patients with low bleeding risk. If
luminal cross-sectional assessment of plaque calcification and
invasive procedures/surgeries are unavoidable during this time,
severity, vessel size, stent apposition, dissection, and thrombus.
dual antiplatelet therapy, or at least aspirin alone, should be
The device is manufactured by both Boston Scientific Corpora-
continued if possible. The complete conclusions and recommen-
tion (Natick, Massachusetts, U.S.) and Volcano Corporation
dations are further detailed in Table 30.4 (46).
(Rancho Cardovo, California, U.S.). When angiographic find-
ings are questionable, IVUS can often provide valuable addi-
Covered Stent tional information. Accepted minimal luminal cross-sectional
The covered stent (Jomed International AB, Helsingborg, areas of 4 and 5 mm2 are used to define critical non–left main
Sweden) features a circumferential polytetrafluoroethylene and left main narrowings, respectively (48,49). The decision
membrane sandwiched between double-stent layers and is regarding use of rotational atherectomy for an angiographically
[26/6/010/22:42:14] [388–403]
calcified lesion may be impacted by superficial versus deep The net clinical outcome was comparable with 13% of patients
location of calcium. Poststent evaluation of strut apposition, reaching the combined ischemic and bleeding end points with
stent-to-vessel sizing, and stent symmetry may indicate the heparin plus GP IIb/IIIa and 12% with bivalirudin monotherapy.
need for further treatment. Finally, IVUS assessment of ISR An interesting substudy showed that patients with major bleed-
may elucidate the mechanism of restenosis, such as inadequate ing episodes had higher mortality at 35 days compared with
stent sizing, plaque prolapse, dissection, or rarely, stent frac- those who did not bleed (51). This observation has led to the
ture, thereby directing therapeutic strategies. inclusion of ischemic events plus bleeding as the primary com-
posite end point in some of the current trials. A similar strategy
was tested in the Harmonizing Outcomes with Revascularization
Pressure Wire and Stents in Acute Myocardial Infarction (HORIZONS-AMI)
The pressure wire features a distal pressure sensor to measure trial, which compared heparin plus a GP IIb/IIIa inhibitor with
the hemodynamic significance of angiographically borderline bivalirudin monotherapy in patients with STEMI. The 30-day
stenoses. This device is manufactured by both Radi Medical outcome showed a net benefit with bivalirudin for ischemic
Systems (Wilmington, Massachusetts, U.S.) and Volcano Cor- and bleeding end points (9.2% vs. 12.1%). Ischemic MACE was
poration. The ratio of pressures distal and proximal to the equivalent, but there was less bleeding with bivalirudin (4.9% vs.
lesion at maximal hyperemia is known as the fractional flow 8.3%). These two trials have resulted in a significant reduction in
reserve (FFR). Adenosine, administered by either intravenous the use of GP IIb/IIIa therapy in both ACS and AMI in favor of
infusion or intracoronary bolus, is often used to induce hyper- bivalirudin therapy.
emia. FFR above and below 0.75 have been validated to corre-
late with functionally benign and significant stenoses,
respectively. Additionally, FFR > 0.90 correlates with adequate Fondaparinux
stent deployment. The Organization for Assessment of Strategies for Ischemic
The FFR offers comprehensive physiologic assessment Syndromes (OASIS)-6 (52) trial reported an increased incidence
and importantly accounts for both the size of the subtended of guiding catheter thrombosis in cases of stand-alone fonda-
territory and the oxygen demand of the myocardium. The same parinux therapy. Hence, concomitant heparin administration
stenosis may be result in either a normal or abnormal FFR with either unfractionated heparin or low-molecular-weight
depending on the adequacy of collateral flow in the subtended heparin is recommended during PCI.
myocardial territory. Additionally, if the perfusion requirement
is low in a predominately scarred region, an anatomically Dual Antiplatelet Therapy
“critical” lesion may be associated with a high FFR, indicating Dual antiplatelet therapy with aspirin and clopidogrel is indi-
an adequate supply-to-demand relationship. FFR accuracy, cated for a minimum of one year and one month after DES and
however, may be limited in situations of high filling pressures BMS implantation, respectively. High-dose aspirin (162–325 mg)
(hypertension or left ventricular hypertrophy), cardiac trans- is recommended for the initial postprocedural period. Addi-
plantation, and saphenous vein grafts. tionally, while the recommended clopidogrel loading dose is
300 mg, higher doses (600 or 900 mg) may be associated with
reduced platelet resistance (53). In one trial, 119 patients under-
ADJUNCT PHARMACOLOGY going DES implantation were randomized to a loading dose of
Heparin and GP IIb/IIIa Vs. Bivalirudin clopidogrel 600 mg followed by 75 mg twice daily for one
Whereas unfractionated heparin has been the traditional anti- month versus a loading dose of 300 mg followed by 75 mg
coagulant used during PCI procedures, recent studies involv- daily. There were no differences in bleeding complications, but
ing more modern agents have challenged its role. Bivalirudin, a high-dose subjects demonstrated greater platelet inhibition (41%
direct thrombin inhibitor, was assessed in the Intracoronary vs. 27%, p ¼ 0.046 at four hours, 19% vs. 10%, p ¼ 0.047 at 30
Stenting and Antithrombotic Regimen—Rapid Early Action for days) and a lower composite end point of cardiovascular death,
Coronary Treatment 3 (ISAR-REACT 3) trial (50). In a double- MI, and TVR (10.3% vs. 23.8%, p ¼ 0.04) (54).
blind fashion, 4570 patients with either stable or unstable Although aspirin resistance has been linked to a clear
angina were randomized to undergo PCI with unfractionated increase in overall MACE, the role of aspirin or clopidogrel
heparin or bivalirudin after a loading dose of clopidogrel 600 mg. resistance in stent thrombosis remains less certain. Part of the
At 30 days, no differences were reported for the composite challenge relates to the lack of definition uniformity for platelet
primary end point of death, MI, urgent TVR, or in-hospital resistance. Additionally, accompanying confounders such as
bleeding. However, major bleeding was reduced in the bivalir- lesion complexity, stent deployment techniques, and patient
udin arm (3.1% vs. 4.6%, RR 0.66, 95% CI, 0.49–0.90, p ¼ 0.008). clinical characteristics have made analyses difficult in the
The Acute Catheterization and Urgent Interventional absence of large randomized trials. Nevertheless, Duzenli and
Triage Strategy (ACUITY) trial compared three strategies for colleagues found that, in diabetic subjects with demonstrated
patients presenting with ACS, the majority of whom went to resistance to aspirin 100 mg daily, increased platelet inhibition
PCI. The anticoagulation arms were unfractionated or low- occurred with both higher aspirin dose (300 mg daily) and dual
molecular weight heparin plus a GP IIb/IIIa inhibitor, bivalir- antiplatelet therapy (clopidogrel 75 mg and aspirin 100 mg
udin plus a GP IIb/IIIa inhibitor, or bivalirudin monotherapy. daily) (54).
The composite primary end point included the ischemic end Clinically significant aspirin resistance has been esti-
points of death, MI, and unplanned revascularization, plus mated to range from 5% to 60% of treated patients, and in
major bleeding. In the PCI group (7789 patients), the ischemic vitro laboratory evidence of such resistance may be associated
end point was 8% with heparin plus GP IIb/IIIa and 9% with with increased cardiovascular event rates (55). Clopidogrel
bivalirudin monotherapy. Major bleeding was reduced from resistance appears dose-dependent; one study reported inci-
7% with heparin and GP IIb/IIIa therapy to 4% with bivalirudin. dences of 28% and 8% for 300- and 600-mg loading doses,
[26/6/010/22:42:14] [388–403]
respectively (56). Furthermore, Buonamici and coauthors found ing during off-hours were less likely to receive PCI (OR 0.93,
clopidogrel resistance to result in a fourfold increased risk of 95% CI, 0.89–0.98), had longer door-to-balloon times (median
DES thrombosis over four-year follow-up (57). A recent evalu- 110 vs. 85 minutes, p < 0.0001), and had lower rates of under
ation of point-of-care platelet inhibition measurement in PCI 90-minute door-to-balloon times (adjusted OR 0.34, 95% CI,
patients reported a sixfold increase in 30-day MACE for sub- 0.29–0.39) (69).
jects with platelet activity in the highest quartile compared with Primary PCI should only be directed at the infarct artery;
the lowest quartile (p ¼ 0.016) (58). The ongoing Gauging treatment of concomitant lesions in other arteries should be
Responsiveness with a VerifyNow Assay—Impact on Throm- staged. Results of the Primary Angioplasty in Myocardial
bosis and Safety (GRAVITAS) trial will evaluate tailored clopi- Infarction (PAMI) trial (70) demonstrated the superiority of
dogrel dosing on the basis of a new bedside platelet assay stenting over balloon angioplasty acutely and up to five years.
system. Furthermore, the Trial to Assess the Use of the Cypher Stent in
Acute Myocardial Infarction Treated with Balloon Angioplasty
(TYPHOON) trial (40) found that SES implantation was safe
SPECIAL CLINICAL SCENARIOS during STEMI and resulted in significantly reduced rates of
STEMI recurrent MI and TVR compared with BMS. However, the
Coronary thrombosis accounts for approximately 70% of car- Paclitaxel-Eluting Stent Versus Conventional Stent in Myocar-
diovascular morbidity and mortality. Plaque rupture, or less dial Infarction with ST-Segment Elevation (PASSION) (71)
frequently plaque erosion, is the substrate for superimposed investigators reported no difference in the primary end point
thrombosis. Other, less common causes of MI include coronary of cardiac death, recurrent MI, or TVR when PES was com-
spasm, spontaneous coronary dissection, coronary embolism, pared with BMS.
coronary vasculitis, and anomalous coronary arteries (59). Although previously controversial, DES implantation
Primary PCI for STEMI remains the only clinical scenario during AMI appears to be safe and efficacious. Mauri and
in which survival benefit from PCI has been unequivocally coinvestigators reviewed 7217 PCI cases, 4016 with DES and
demonstrated. 3201 with BMS. Utilizing propensity-score matching, they
found superior two-year risk-adjusted mortality rates for
Pathology/Pathophysiology of STEMI patients treated with DES versus BMS (10.7% vs. 12.8%, p ¼
Over two centuries ago, Virchow first described arterial plaque 0.02). The DES advantage was present for STEMI (8.5% vs.
rupture and analogized the necrotic core to that of an abscess. 11.6%, p ¼ 0.008) and non-STEMI (12.8% vs. 15.6%, p ¼ 0.04)
Rupture of thin-cap fibroatheromas is the most common path- subgroups. Moreover, recurrent MI rates were reduced for non-
ophysiology of STEMI (59). In their sudden death autopsy STEMI patients treated with DES; and repeat revascularization
series, Virmani et al. found that 65% to 70% of atherothrombi rates were superior for DES patients in all groups. The authors
involved plaque rupture, whereas 25% to 30% arose from concluded that DES implantation resulted in superior two-year
plaque erosion (60). These etiologies are gender-specific; plaque mortality and TVR compared with BMS (13). Although these
erosion occurs with twice the frequency in women versus men results are encouraging, they may be influenced by selection
(61). However, this phenomenon is rarely observed in preme- bias and use of prolonged dual antiplatelet therapy (72). This is
nopausal women (62). supported by the HORIZONS-AMI trial that showed no differ-
ence in death, MI or stroke rates between DES and BMS (73).
Pharmacology Fragmentation of a friable thrombus can lead to distal
The enhanced thrombotic milieu necessitates potent antithrom- microembolization during primary PCI. Angiographic slow
botic and antiplatelet therapy; thus, STEMI patients are at flow or no reflow represents microembolization of fibrin and
increased risk for bleeding. Significant bleeding, as well as platelets into the arteriolar bed. Selection of slightly longer stent
the need for transfusion, has been found to predict one-year length may help “trap” much of the extruded material.
mortality (63). The HORIZONS-AMI trial (64) demonstrated Various distal protection devices have been designed to
lower composite end points and major bleeding for bivalirudin trap and subsequently remove debris after stent placement.
compared with heparin plus a GP IIb/IIIa inhibitor. Addition- Aspiration devices physically extract the friable components of
ally, while overall MACE was equivalent, 30-day mortality was the plaque. Pretreatment of these high-risk lesions with manual
significantly less with bivalirudin (2.1% vs. 3.1%, p ¼ 0.047). Of catheter aspiration has recently been demonstrated in the
note, the rate of acute thrombosis (<24 hours, ARC definite or Thrombus Aspiration During Percutaneous Coronary Interven-
probable thrombosis) was higher with bivalirudin, although the tion in Acute Myocardial Infarction Study (TAPAS) (13) to
overall 30-day stent thrombosis rates were similar. The ACC/ confer acute and one-year benefit. At one year, cardiac death
AHA have accordingly assigned class IIA status to concomitant (3.6% vs. 6.7%, p ¼ 0.020) and cardiac death or nonfatal
clopidogrel administration with upstream bivalirudin (65). reinfarction (5.6% vs. 9.9%, p ¼ 0.009) rates were lower with
thrombus aspiration compared with PCI only.
When no reflow is encountered, intracoronary adminis-
Technical Considerations tration of vasodilators such as nitroprusside, adenosine, or
Multiple randomized trials have demonstrated the superiority calcium-channel blockers through the central lumen of an
of primary angioplasty over fibrinolytic therapy in STEMI over-the-wire balloon catheter into the distal vascular bed
(66–68). The door-to-balloon time should be within 90 minutes. frequently provides prompt resolution. It is important to bear
A triage system with immediate notification of the on-call in mind that surgical revascularization offers no benefit in cases
interventionalist and catheterization laboratory staff by the of refractory no reflow. The over-the-wire system is further
emergency department facilitates this process. Nonetheless, useful in distal visualization of a totally occluded artery. After
many eligible patients still fail to receive optimal care. A recent initial guidewire lesion penetration, the balloon can be
analysis of 20,279 STEMI patients found that patients present- advanced distally then retracted, without inflation. The slightly
[26/6/010/22:42:14] [388–403]
higher balloon profile often results in a sufficient channel to Right heart catheterization is invaluable in diagnosing
allow visualization of the distal coronary artery. If distal flow is and monitoring these patients. Constant pulmonary arterial
still not achieved at this point, the balloon is then advanced pressure and frequent cardiac index assessments guide titration
distally followed by removal of the guidewire. Contrast of inotropic and pressor agents. Aside from aggressive phar-
medium is subsequently injected through the central balloon macologic and ventilatory support measures, emergent angiog-
lumen to opacify the distal artery, confirming intraluminal raphy and revascularization, as well as mechanical circulatory
position. devices, are warranted. The intra-aortic balloon pump provides
In most cases of STEMI, direct stenting (without balloon augmentation of diastolic coronary perfusion, cardiac index,
before dilatation) may be preferable, as aggressive balloon and mean arterial blood pressure. Data from the National
inflations can result in lesion fragmentation. Use of smaller Registry of Myocardial Infarction revealed significant survival
diameter balloons (2.0 or 2.5 mm) sufficient to allow stent advantage associated with intra-aortic balloon counterpulsa-
passage is recommended. Additionally, a single high-pressure tion (77). Intra-aortic counterpulsation is contraindicated in
stent inflation may minimize the “cheese-grating” extrusion cases of significant aortic insufficiency and aortic dissection.
and subsequent downstream embolization of friable thrombus. Total circulatory support may be needed in refractory
cases. Left ventricular assist devices drain blood from the left
heart and pump it back into the systemic arterial circulation.
Cardiogenic Shock The TandemHeart device (Cardiac Assist, Inc., Pittsburgh,
The occurrence of cardiogenic shock has notably declined in Pennsylvania, U.S.) is a percutaneously inserted device consist-
recent years, coincident with the increasingly widespread prac- ing of a transeptal left atrial inflow catheter and a femoral
tice of primary PCI for STEMI. This dreaded complication of artery outflow catheter. The Impella device (Abiomed, Inc.,
AMI occurs in 5% to 8% and 2.5% of all patients with STEMI Danvers, Massachusetts, U.S.) operates on a similar principle,
and non-STEMI, respectively. In-hospital mortality is approxi- but is inserted across the aortic valve. Extracorporeal circula-
mately 50%. Clinical risk factors for development of cardio- tory devices additionally pass the venous blood through a
genic shock include anterior STEMI, left bundle-branch block, membrane oxygenator, thereby incorporating oxygenation
older age, multivessel CAD, and history of heart failure, dia- with perfusion support. Although left ventricular assist devices
betes, hypertension, prior MI or angina (74). Although com- offer superior hemodynamic support over intra-aortic balloon
monly diagnosed clinically by symptoms or signs of left counterpulsation, current data fail to demonstrate a survival
ventricular or biventricular failure, formal criteria for cardio- difference (78).
genic shock include marked hypotension (systolic pressure < 80 Urgent revascularization was shown to impart a 13%
mmHg or reduction of baseline mean arterial pressure by 30 absolute increase in survival for patients in the SHOCK trial
mmHg) and pump failure (cardiac index < 1.8 L/min m2) in (NNT < 8) (76). The ACC/AHA guidelines assign class I and
the presence of adequate filling pressures (LV end-diastolic class II indications to urgent revascularization for cardiogenic
pressure > 18 mmHg) (74). shock in patients younger and older than 75 years, respectively.
Mechanical complications occur infrequently, but usually Adherence to the 90-minute door-to-balloon time is less impor-
have dramatic presentations. Acute mitral regurgitation results tant in the setting of shock, with survival benefit demonstrated
from papillary muscle ischemia or rupture, and acute ventric- up to 48 hours after AMI and 18 hours after shock onset (74). If
ular septal rupture can result in a large left-to-right shunt. significant delay to primary PCI is anticipated, fibrinolytic
These two entities have clinically similar presentations of therapy can be administered, although the scope of benefit is
marked pulmonary congestion, hypotension, and holosystolic clearly diminished. Multivessel disease can be treated with
murmurs. The harsh and turbulent flow across the interven- either CABG or PCI. Two-thirds of the patients who underwent
tricular septum may be associated with a palpable thrill along PCI in the SHOCK trial had multivessel disease (76). A recent
the sternal border, whereas most mitral regurgitant murmurs follow-up study of the SHOCK registry reported promising
are not. Echocardiography and/or sequential blood oxygen three- and six-year survival rates of 41.4% and 32.8% for
saturations obtained from the right atrium and pulmonary patients receiving early PCI (79). Primary PCI for cardiogenic
artery (“sat run”) can frequently distinguish these etiologies. shock is further a separate chapter.
Additionally, while free ventricular wall rupture, resulting in Isolated or predominant right ventricular infarction
immediate tamponade, is almost universally fatal, a contained accounts for 5% of cardiogenic shock. The mainstay of initial
rupture may allow time for urgent diagnosis and intervention. therapy consists of volume resuscitation and inotropic support.
For all mechanical causes of cardiogenic shock, emergent sur- Not infrequently, left ventricular function may also be compro-
gical repair is crucial and potentially life-saving. mised by impaired filling due to leftward shift of the interven-
Most commonly, however, cardiogenic shock results tricular septum as well as decreased preload (80).
from significant left ventricular dysfunction. While takotsubo
cardiomyopathy (also known as the “broken heart” or “apical
ballooning” syndrome) (75) has become an increasingly recog- BIFURCATION LESIONS
nized etiology of STEMI and cardiogenic shock, the vast major- Despite dramatic improvements in PCI techniques and equip-
ity of cardiogenic shock occurs in the setting of coronary artery ment, bifurcation lesions remain a serious challenge for the
occlusion and STEMI. Of note, the degree of left ventricular modern interventionalist. Turbulent flow dynamics and high
dysfunction may not be predictive of clinical presentation. In shear stress likely predispose to plaque formation in such
the Should We Emergently Revascularize Occluded Coronaries locations. These lesions comprise 15% to 20% of PCI. Compared
for Cardiogenic Shock (SHOCK) trial (76), Hochman and asso- with other interventions, bifurcation PCI is associated with
ciates reported a mean left ventricular ejection fraction of lower procedural success rates, higher procedural costs, longer
almost 30%. Nonetheless, left ventricular ejection fraction hospitalizations, and higher clinical and angiographic resteno-
remains a strong predictor of long-term prognosis (74). sis rates. Some studies have reported bifurcation lesions as an
[26/6/010/22:42:14] [388–403]
independent risk factor for subacute and late stent thrombosis The Medina classification (85) is a simplified, practical
(81,82). Compared with BMS, DES are associated with less TVR system involving binary allocation of “1” or “0” based on the
and main branch restenosis in this anatomic subset (1). respective presence or absence of significant plaque burden in
the proximal main branch, distal main branch, or side branch.
A “1,1,0” bifurcation lesion, for instance, would indicate prox-
Lesion Classification imal and distal main branch involvement, with sparing of the
The presence of atheromatous plaque causing >50% stenosis in side branch (Fig. 30.5). The location of the bifurcation lesion
both the main and side branches is considered the definition of relative to the carina is a crucial determinant in the decision
a true bifurcation lesion. Several classification schema of vary- regarding a single- or double-stent strategy.
ing complexity have been proposed. The Duke classification
(83) is ordered A through F according to main branch and side
branch plaque location (Fig. 30.3). Techniques
Lefevre et al. (84) have proposed an alternate classifica- Debulking Techniques
tion based on the side branch to main branch angle and lesion Plaque shift or “snow-plowing” can occur in 4.5% to 26% of
morphology; bifurcation lesions are classified as Y-shaped or T- bifurcation lesion PCI (86,87) and is more frequent with smaller
shaped (side branch to main branch angle less than or greater side branch reference diameter, side branch origination from
than 708, respectively). In most cases, Y-shaped lesions allow the main branch lesion, and ACS presentation (87). Occlusion of
easier side branch access but may be associated with greater the side branch often results in periprocedural biomarker ele-
plaque shift. The converse is generally true for T-shaped vation, which may negatively impact long-term MACE, includ-
lesions. By this scheme, type 1 lesions are defined as true ing TVR (86). Fortunately, more than 75% of side branch
bifurcation lesions involving the main branch, proximal and occlusions are patent at angiographic follow-up (88).
distal to the bifurcation, as well as the side branch ostium. Type Prestent debulking strategies utilizing rotational, direc-
2 lesions involve only the main branch at the bifurcation, tional, or laser atherectomy have been proposed to minimize
sparing the side branch ostium. Type 3 lesions are located in potential plaque shift. The FLEXI-CUT (89) study evaluated
the main branch proximal to the bifurcation. Type 4 lesions directional coronary atherectomy-assisted single-vessel stent-
involve only the bifurcation, without proximal or distal disease ing of left main bifurcation lesions. The procedural success rate
in either branch. The latter is further divided into subtypes 4A was 96.7%, with 10.3% TVR and 6.9% overall MACE rates.
and 4B, isolated to the main branch and side branch, respec- Interestingly, Karvouni and colleagues found that prestent
tively (Fig. 30.4). directional atherectomy resulted in a lower procedural success
Figure 30.3 The Duke bifurcation lesion classifica-

tion. Source: From Ref. 83.
[26/6/010/22:42:14] [388–403]
Figure 30.4 The Lefevre bifurcation lesion classification. Source: From Ref. 84.
Stent Techniques
While expert opinion varies regarding the preferred technique
for treating bifurcation lesions, employment of at least one DES
is recommended in most cases. When the risk for compromis-
ing a large side branch is low, main branch stenting with
provisional side branch stenting frequently offers the simplest
and most practical solution. The Nordic Bifurcation Study (90)
was a DES study comparing the results of two-vessel stenting
versus main branch stenting with provisional side branch
stenting. At six months, there was no difference in MACE;
however, two-vessel stenting was associated with significantly
greater procedural and fluoroscopic times, contrast volumes,
and rates of postprocedural biomarker elevation.
In cases requiring double-stent implantation, many tech-
niques have been proposed. The T-stent technique (91) involves
stenting the side branch ostium after initial stent deployment in
the main branch. A second wire is initially left in the side
branch and jailed by the main branch stent. The favorable side
branch angle modification and angiographic reference offered
by the jailed wire facilitates subsequent side branch access
through the deployed stent using the main branch wire. After
Figure 30.5 The Medina bifurcation lesion classification. Source:
dilating the stent struts, a second stent is meticulously
From Ref. 85.
implanted at the side branch ostium with a final kissing balloon
dilatation. The technique is best suited to bifurcation lesions
with an angle close to or equal to 908. Nonetheless, incomplete
coverage of the side branch ostium remains a major limitation.
rate (87.1% vs. 100%, p ¼ 0.03), due to higher non–Q wave MI In the modified T-stenting technique, the side branch
rates (12.9% vs. 0%, p ¼ 0.03). However, directional atherec- stent is advanced into position followed by main branch stent
tomy-stenting was associated with greater acute gain, as well as positioning without deployment. The former is then carefully
a trend toward reduced MACE at follow-up (p ¼ NS) (17). positioned at the ostium and deployed first with only the
Nonetheless, because of the relative technical complexity of proximal stent marker protruding into the main branch. Fol-
these approaches, single- or double-vessel stent strategies lowing removal of delivery catheter and guidewire from the
remain preferred in the majority of cases. side branch, the main branch stent is then deployed across the
[26/6/010/22:42:14] [388–403]
ostium of the side branch. The side branch is then rewired for lesion coverage but is limited by its procedural complexity and
performance of a final kissing balloon dilatation. the need for three stents.
The crush technique (92) ensures complete lesion cover-
age of the side branch ostium. As in the modified T-stenting
technique, the side branch stent is positioned and deployed LIMITATIONS OF PCI
first. However, in this case, the proximal side branch stent PCI Vs. Medical Therapy
marker is placed 2 to 3 mm proximal to the bifurcation within Despite widespread popularity, PCI has not been shown to
the main branch. After withdrawal of the side branch wire and decrease mortality in patients with stable CAD. Conversely,
stent delivery catheter, the main branch stent is subsequently primary PCI provides a survival advantage compared with
inflated, crushing the proximal segment of side branch stent. fibrinolytic therapy in patients with STEMI.
Three layers of stent struts are thus compressed against the In the Clinical Outcomes Utilizing Revascularization and
ipsilateral main branch wall. Finally, kissing balloon dilatation Aggressive Drug Evaluation (COURAGE) trial (96) over 35,000
is undertaken (through the triple layer of stent struts at the patients with chronic stable angina and angiographically-
bifurcation). This last step is essential to ensure adequate documented CAD were screened for randomization to optimal
carinal and side branch ostial patency and to reduce long- medical therapy with or without PCI. Inclusion criteria
term TVR of either branch. included >70% stenosis in a proximal major epicardial artery,
In their IVUS follow-up of DES crush technique in 40 CCS class I to III angina, and precatheterization objective evi-
subjects, Costa and colleagues (93) found that main branch dence of ischemia. Patients with uncontrolled angina, compli-
stent minimum luminal diameter measured <4 and <5 mm2 in cated post-MI course, EF < 30%, or cardiogenic shock were
8% and 20% of lesions, respectively. For the side branch ostium, excluded. Only 2287 patients (<10%) were enrolled (1149 opti-
minimal stent areas of <4 and <5 mm2 were seen in 44% and mal medical therapy and 1138 PCI). Few patients received DES
76%, respectively. Incomplete apposition of side branch or in this trial. No difference was observed for the primary com-
main branch stent struts against the precarinal main branch posite end point of death or MI at a median of 4.6-year follow-
wall was seen in the majority (>60%) of non–left main lesions. up (96). Subsequent revascularization for ischemia was
Consistent with other reports, the minimal side branch area required in 21% PCI and 32.6% medical patients.
was observed at the ostium, suggesting this as a likely contrib- The 314-patient nuclear stress test substudy reported
utor to the higher restenosis rate at this location. significantly greater ischemic burden reductions with PCI ver-
The culotte (or “pant-leg”) technique (94) ensures com- sus optimal medical therapy alone (78% vs. 52%, p ¼ 0.007).
plete bifurcation carinal coverage. In its original description, Additionally, those patients deriving significant ischemia
after double-wire insertion, the first stent is deployed in the improvement also demonstrated lower unadjusted mortality
more sharply angulated artery (usually the side branch) at high or MI rates (p ¼ 0.037). The unadjusted primary end point
pressure (12–14 atm). The main branch wire is thus trapped, reduction was especially pronounced in those individuals with
and a third wire is introduced to recross the struts of the initial moderate to severe baseline ischemia (p ¼ 0.001). While DES
stent into the other branch. After removal of the jailed wire, use would not likely have impacted mortality results, it prob-
balloon predilatation through the stent struts facilitates passage ably would have decreased TVR and recurrent angina rates. It
of a second stent into the other vessel. The procedure ends with would thus seem reasonable that patients with CAD and
a final kissing inflation. Technical complexity and high rates of moderate to severe ischemia should be considered for PCI as
postprocedural events and restenosis in registry reports have an adjunct to optimal medical therapy.
limited the popularity of this strategy. The Medicine, Angioplasty, or Surgery Study (MASS) II
The simultaneous kissing stent (94) technique is usually compared medical therapy, PCI, and CABG in patients with
employed for bifurcation lesions with similar-sized branches stable multivessel CAD (97). At five years, no significant differ-
and large proximal vessel reference diameter. The procedure ences were observed in subsequent revascularization and over-
involves wiring the main and side branches and allows main- all MACE rates between medical and PCI patients.
tenance of access to both during the entire procedure. Both Conversely, the Japanese Stable Angina Pectoris (JSAP)
stents are positioned side by side and sequentially inflated, study reported a possible benefit for PCI/optimal medical
followed by a final kissing inflation. A “double-barrel” config- therapy versus optimal medical therapy alone in 384 low-risk
uration results, and the carina is thus extended proximally. The patients with stable angina and single- or double-vessel CAD
simultaneous stent deployment minimizes plaque shift. The (98). Over a 3.3-year follow-up, the PCI group demonstrated a
advantage of this technique lies in its ease and predictability. significantly reduced combined death and ACS risk (7.9% vs.
Kim and colleagues (95) reported a procedural success rate of 14.9%, p ¼ 0.018). Similarly, combined death, ACS, and stroke
100%. At 26-month follow-up, there were no deaths, MI, or (8.5% vs. 16.4%, p ¼ 0.044) and combined death, ACS, stroke,
stent thromboses. TVR occurred in five patients (14%), with and emergency hospitalization (22% vs. 33.2%, p ¼ 0.04) rates
four (13%) in the main branch and three (10%) in the side were decreased for PCI patients.
branch. Of note, in 14 patients (47%), a membranous film
coating the double-layered carina was noted both angiograph- Left Main and Multivessel CAD
ically and by IVUS. However, no clinical sequelae were asso- While CABG has traditionally been recommended for patients
ciated with this membrane. with triple-vessel and left main CAD, the enhanced safety and
The V-stent variant involves proximal alignment of both restenosis benefits of PCI due to technical and device improve-
stents at the carina. While this strategy minimizes proximal ments have prompted randomized trials comparing PCI and
main branch dissection, carinal coverage may not be assured. CABG. No significant mortality or MI differences have been
Y-stenting is performed by additional deployment of a proxi- demonstrated, although TVR was greater in PCI patients (99).
mal main branch stent juxtaposed against the proximal edges of The recent Synergy Between Percutaneous Coronary
both V stents. This technique was developed to allow complete Intervention with Taxus and Cardiac Surgery (SYNTAX) trial
[26/6/010/22:42:14] [388–403]
randomized patients with left main or triple-vessel CAD to 8. Ryan TJ, Faxon DP, Gunnar RM, et al. Guidelines for percuta-
multivessel PCI or CABG (100). The study protocol was reflec- neous transluminal coronary angioplasty. A report of the Amer-
tive of modern PCI strategies using DES and optimal CABG ican College of Cardiology/American Heart Association Task
techniques, with overall and bilateral mammary arterial con- Force on Assessment of Diagnostic and Therapeutic Cardiovas-
cular Procedures (Subcommittee on Percutaneous Transluminal
duit rates of 97% and 27.6%, respectively. Additionally, coro-
Coronary Angioplasty). Circulation 1988; 78:486–502.
nary anatomy was complex with 84% bifurcation or trifurcation 9. Krone RJ, Laskey WK, Johnson C, et al. A simplified lesion
lesions, 22% chronic total occlusions, and 39% significant left classification for predicting success and complications of coro-
main disease. A mean of 4.6 DES were implanted per patient nary angioplasty. Registry Committee of the Society for Cardiac
with a mean total stent length of 86 mm (one-third of patients Angiography and Intervention. Am J Cardiol 2000; 85:1179–1184.
>100 mm). 10. Fumiaki I, Yeung AC. Equipment for PCI. In: King SB III, ed.
There was a higher MACCE (major adverse cardiovascu- Interventional Cardiology. New York: McGraw Hill Medical,
lar or cerebrovascular event) rate for PCI (17.8% vs. 12.1% for 2007:79–86.
CABG, p ¼ 0.0015). Hence, the noninferiority primary end point 11. Vlaar PJ, Svilaas T, van der Horst IC, et al. Cardiac death and
for PCI was not achieved. Consistent with prior findings, reinfarction after 1 year in the Thrombus Aspiration during Percu-
taneous coronary intervention in Acute myocardial infarction Study
combined rates for all-cause mortality, MI, and stroke were
(TAPAS): a 1-year follow-up study. Lancet 2008; 371: 1915–1920.
equivalent between groups (7.6% for PCI and 7.7% CABG, p ¼ 12. Ali A, Cox D, Dib N, et al. Rheolytic thrombectomy with
0.98); and TVR after PCI exceeded that of CABG by 7.8%. percutaneous coronary intervention for infarct size reduction in
However, a significantly higher stroke risk was observed with acute myocardial infarction: 30-day results from a multicenter
CABG (2.2% vs. 0.6%, p ¼ 0.003). Despite challenging coronary randomized study. J Am Coll Cardiol 2006; 48:244–252.
anatomy, excess TVR with PCI was much less than that 13. Mauri L, Kinlay S. Adjunctive devices: atherectomy, thrombec-
observed in previous PCI versus CABG trials. Interestingly, tomy, ebolic protection, IVUS, Doppler and pressure wires. In:
rates of symptomatic bypass graft occlusion and stent throm- King SB III, ed. Interventional Cardiology. New York: McGraw
bosis were equivalent. Hill Medical, 2007:93–106.
14. Chen JP. Dual lumen catheters: more than just aspiration.
J Invasive Cardiol 2006; 18:346.
CONCLUSION 15. Margheri M, Vittori G, Chechi T, et al. Thrombus aspiration with
export catheter in ST elevation myocardial infarction. J Interv
Despite recent advances in treating structural heart disease, PCI
Cardiol 2007; 20:38–43.
remains the cornerstone of interventional cardiology; and tech- 16. Baim DS, Wahr D, George B, et al. Randomized trial of a distal
nical improvements have dramatically improved outcomes. A embolic protection device during percutaneous intervention of
continuing challenge will be appropriate use of PCI (101). In saphenous vein aorto-coronary bypass grafts. Circulation 2002;
addition, optimal medical therapy must be applied to achieve 105:1285–1290.
optimal long-term outcomes for PCI patients. 17. Stone GW, Rogers C, Hermiller J, et al. Randomized comparison
of distal protection with a filter-based catheter and a balloon
occlusion and aspiration system during percutaneous interven-
REFERENCES tion of diseased saphenous vein aorto-coronary bypass grafts.
1. King SB III, Smith SC Jr., Hirshfeld JW Jr., et al. 2007 focused Circulation 2003; 108:548–553.
update of the ACC/AHA/SCAI 2005 guideline update for 18. Bittl JA. Role of adjunctive devices: cutting balloon, thrombec-
percutaneous coronary intervention: a report of the American tomy, laser, ultrasound, and atherectomy. In: Topol EJ, ed. Text-
College of Cardiology/American Heart Association Task Force book of Interventional Cardiology. 5th ed. Philadelphia, PA:
on Practice Guidelines: 2007 writing group to review new evi- Saunders Elsevier, 2008:619–639.
dence and update the ACC/AHA/SCAI 2005 guideline update 19. Whitlow PL, Bass TA, Kipperman RM, et al. Results of the study
for percutaneous coronary intervention, writing on behalf of the to determine rotablator and transluminal angioplasty strategy
2005 writing committee. Circulation 2008; 117:261–295. (STRATAS). Am J Cardiol 2001; 87:699–705.
2. Antman EM, Cohen M, Bernink PJ, et al. The TIMI risk score for 20. vom Dahl J, Dietz U, Haager PK, et al. Rotational atherectomy
unstable angina/non-ST elevation MI: a method for prognosti- does not reduce recurrent in-stent restenosis: results of the angio-
cation and therapeutic decision making. JAMA 2000; 284: plasty versus rotational atherectomy for treatment of diffuse in-
835–842. stent restenosis trial (ARTIST). Circulation 2002; 105:583–588.
3. Granger CB, Goldberg RJ, Dabbous O, et al. Predictors of hospital 21. Topol EJ, Leya F, Pinkerton CA, et al. A comparison of direc-
mortality in the global registry of acute coronary events. Arch tional atherectomy with coronary angioplasty in patients with
Intern Med 2003; 163:2345–2353. coronary artery disease. The CAVEAT Study Group. N Engl J
4. Mehta SR, Cannon CP, Fox KA, et al. Routine vs. selective Med 1993; 329:221–227.
invasive strategies in patients with acute coronary syndromes: 22. Adelman AG, Cohen EA, Kimball BP, et al. A comparison of
a collaborative meta-analysis of randomized trials. JAMA 2005; directional atherectomy with balloon angioplasty for lesions of
293:2908–2917. the left anterior descending coronary artery. N Engl J Med 1993; 329:
5. de Winter RJ, Windhausen F, Cornel JH, et al. Early invasive 228–233.
versus selectively invasive management for acute coronary syn- 23. Holmes DR Jr., Topol EJ, Califf RM, et al. A multicenter,
dromes. N Engl J Med 2005; 353:1095–1104. randomized trial of coronary angioplasty versus directional
6. Invasive compared with non-invasive treatment in unstable cor- atherectomy for patients with saphenous vein bypass graft
onary-artery disease: FRISC II prospective randomised multi- lesions. CAVEAT-II Investigators. Circulation 1995; 91:1966–1974.
centre study. FRagmin and Fast Revascularisation during 24. Stankovic G, Colombo A, Bersin R, et al. Comparison of direc-
InStability in Coronary artery disease Investigators. Lancet 1999; tional coronary atherectomy and stenting versus stenting alone
354:708–715. for the treatment of de novo and restenotic coronary artery
7. Cannon CP, Weintraub WS, Demopoulos LA, et al. Comparison narrowing. Am J Cardiol 2004; 93:953–958.
of early invasive and conservative strategies in patients with 25. Deckelbaum LI, Natarajan MK, Bittl JA, et al. Effect of intra-
unstable coronary syndromes treated with the glycoprotein coronary saline infusion on dissection during excimer laser
IIb/IIIa inhibitor tirofiban. N Engl J Med 2001; 344:1879–1887. coronary angioplasty: a randomized trial. The Percutaneous
[26/6/010/22:42:14] [388–403]
Excimer Laser Coronary Angioplasty (PELCA) Investigators. 44. Chen JP, Crisco LV, Jabara R, et al. Late angiographic stent
J Am Coll Cardiol 1995; 26:1264–1269. thrombosis: the LAST straw for drug-eluting stents? Angiology
26. Reifart N, Vandormael M, Krajcar M, et al. Randomized compar- 2008; 59:667–675.
ison of angioplasty of complex coronary lesions at a single center. 45. Grines CL, Bonow RO, Casey DE Jr., et al. Prevention of prema-
Excimer Laser, Rotational Atherectomy, and Balloon Angioplasty ture discontinuation of dual antiplatelet therapy in patients with
Comparison (ERBAC) Study. Circulation 1997; 96:91–98. coronary artery stents: a science advisory from the American
27. Appelman YE, Piek JJ, Strikwerda S, et al. Randomised trial of Heart Association, American College of Cardiology, Society for
excimer laser angioplasty versus balloon angioplasty for treatment Cardiovascular Angiography and Interventions, American Col-
of obstructive coronary artery disease. Lancet 1996; 347:79–84. lege of Surgeons, and American Dental Association, with repre-
28. U.S. Food and Drug Administration. Update to FDA statement sentation from the American College of Physicians. Circulation
on coronary drug-eluting stents, 2007. Available at: http://www. 2007; 115:813–818.
accessdata.fda.gov/scripts/cdrh/cfdocs/medsun/news/printer. 46. Stankovic G, Colombo A, Presbitero P, et al. Randomized eval-
cfm?id=575. Accessed on January 10, 2009. uation of polytetrafluoroethylene-covered stent in saphenous
29. Morice MC, Serruys PW, Sousa JE, et al. A randomized compar- vein grafts: the Randomized Evaluation of polytetrafluoroethy-
ison of a sirolimus-eluting stent with a standard stent for coro- lene COVERed stent in Saphenous vein grafts (RECOVERS)
nary revascularization. N Engl J Med 2002; 346:1773–1780. Trial. Circulation 2003; 108:37–42.
30. Moses JW, Leon MB, Popma JJ, et al. Sirolimus-eluting stents 47. Schachinger V, Hamm CW, Munzel T, et al. A randomized trial
versus standard stents in patients with stenosis in a native of polytetrafluoroethylene-membrane-covered stents compared
coronary artery. N Engl J Med 2003; 349:1315–1323. with conventional stents in aortocoronary saphenous vein grafts.
31. Stone GW, Ellis SG, Cox DA, et al. A polymer-based, paclitaxel- J Am Coll Cardiol 2003; 42:1360–1369.
eluting stent in patients with coronary artery disease. N Engl J 48. Abizaid AS, Mintz GS, Mehran R, et al. Long-term follow-up
Med 2004; 350:221–231. after percutaneous transluminal coronary angioplasty was not
32. Serruys PW, Ong AT, van Herwerden LA, et al. Five-year out- performed based on intravascular ultrasound findings: impor-
comes after coronary stenting versus bypass surgery for the tance of lumen dimensions. Circulation 1999; 100:256–261.
treatment of multivessel disease: the final analysis of the Arterial 49. Jasti V, Ivan E, Yalamanchili V, et al. Correlations between
Revascularization Therapies Study (ARTS) randomized trial. fractional flow reserve and intravascular ultrasound in patients
J Am Coll Cardiol 2005; 46:575–581. with an ambiguous left main coronary artery stenosis. Circula-
33. Camenzind E, Steg PG, Wijns W. Stent thrombosis late after tion 2004; 110:2831–2836.
implantation of first-generation drug-eluting stents: a cause for 50. Kastrati A, Neumann FJ, Mehilli J, et al. Bivalirudin versus
concern. Circulation 2007; 115:1440–1455; discussion 55. unfractionated heparin during percutaneous coronary interven-
34. Cutlip DE, Windecker S, Mehran R, et al. Clinical end points in tion. N Engl J Med 2008; 359:688–696.
coronary stent trials: a case for standardized definitions. Circu- 51. Manoukian SV, Feit F, Mehran R, et al. Impact of major bleeding
lation 2007; 115:2344–2351. on 30-day mortality and clinical outcomes in patients with acute
35. Pfisterer M, Brunner-La Rocca HP, Buser PT, et al. Late clinical coronary syndromes: an analysis from the ACUITY Trial. J Am
events after clopidogrel discontinuation may limit the benefit of Coll Cardiol 2007; 49:1362–1368.
drug-eluting stents: an observational study of drug-eluting ver- 52. Yusuf S, Mehta SR, Chrolavicius S, et al. Effects of fondaparinux
sus bare-metal stents. J Am Coll Cardiol 2006; 48:2584–2591. on mortality and reinfarction in patients with acute ST-segment
36. Wenaweser P, Daemen J, Zwahlen M, et al. Incidence and elevation myocardial infarction: the OASIS-6 randomized trial.
correlates of drug-eluting stent thrombosis in routine clinical JAMA 2006; 295:1519–1530.
practice. 4-year results from a large 2-institutional cohort study. J 53. von Beckerath N, Taubert D, Pogatsa-Murray G, et al. Absorp-
Am Coll Cardiol 2008; 52:1134–1140. tion, metabolization, and antiplatelet effects of 300-, 600-, and
37. Fajadet J, Wijns W, Laarman GJ, et al. Randomized, double-blind, 900-mg loading doses of clopidogrel: results of the ISAR-
multicenter study of the Endeavor zotarolimus-eluting phos- CHOICE (Intracoronary Stenting and Antithrombotic Regimen:
phorylcholine-encapsulated stent for treatment of native coro- Choose Between 3 High Oral Doses for Immediate Clopidogrel
nary artery lesions: clinical and angiographic results of the Effect) Trial. Circulation 2005; 112:2946–2950.
ENDEAVOR II trial. Circulation 2006; 114:798–806. 54. Abuzahra M, Pillai M, Caldera A, et al. Comparison of higher
38. Frobert O, Lagerqvist B, Carlsson J, et al. Differences in restenosis clopidogrel loading and maintenance dose to standard dose on
rate with different drug-eluting stents in patients with and platelet function and outcomes after percutaneous coronary
without diabetes mellitus: a report from the SCAAR (Swedish intervention using drug-eluting stents. Am J Cardiol 2008; 102:
Angiography and Angioplasty Registry). J Am Coll Cardiol 2009; 401–403.
53:1660–1667. 55. Snoep JD, Hovens MM, Eikenboom JC, et al. Association of
39. Kastrati A, Mehilli J, Pache J, et al. Analysis of 14 trials compar- laboratory-defined aspirin resistance with a higher risk of recur-
ing sirolimus-eluting stents with bare-metal stents. N Engl J Med rent cardiovascular events: a systematic review and meta-analy-
2007; 356:1030–1039. sis. Arch Intern Med 2007; 167:1593–1599.
40. Spaulding C, Daemen J, Boersma E, et al. A pooled analysis of 56. Gurbel PA, Bliden KP, Hayes KM, et al. The relation of dosing to
data comparing sirolimus-eluting stents with bare-metal stents. clopidogrel responsiveness and the incidence of high post-treat-
N Engl J Med 2007; 356:989–997. ment platelet aggregation in patients undergoing coronary stent-
41. Stone GW, Moses JW, Ellis SG, et al. Safety and efficacy of ing. J Am Coll Cardiol 2005; 45:1392–1396.
sirolimus- and paclitaxel-eluting coronary stents. N Engl J Med 57. Buonamici P, Marcucci R, Migliorini A, et al. Impact of platelet
2007; 356:998–1008. reactivity after clopidogrel administration on drug-eluting stent
42. Stettler C, Wandel S, Allemann S, et al. Outcomes associated with thrombosis. J Am Coll Cardiol 2007; 49:2312–2317.
drug-eluting and bare-metal stents: a collaborative network 58. Patti G, Nusca A, Mangiacapra F, et al. Point-of-care measure-
meta-analysis. Lancet 2007; 370:937–948. ment of clopidogrel responsiveness predicts clinical outcome in
43. Patti G, Nusca A, Di Sciascio G. Meta-analysis comparison (nine patients undergoing percutaneous coronary intervention results
trials) of outcomes with drug-eluting stents versus bare metal of the ARMYDA-PRO (Antiplatelet therapy for Reduction of
stents in patients with diabetes mellitus. Am J Cardiol 2008; MYocardial Damage during Angioplasty-Platelet Reactivity Pre-
102:1328–1334. dicts Outcome) study. J Am Coll Cardiol 2008; 52:1128–1133.
[26/6/010/22:42:14] [388–403]
59. Burke AP, Ladich E, Kutys R, et al. The thrombotic AMI lesson: 78. Burkhoff D, Cohen H, Brunckhorst C, et al. A randomized
lessons from pathology. In: Mehta SR, ed. Textbook of STEMI multicenter clinical study to evaluate the safety and efficacy of
Interventions. Malvern: HMP Communications, 2008:3–12. the TandemHeart percutaneous ventricular assist device versus
60. Virmani R, Kolodgie FD, Burke AP, et al. Lessons from sudden conventional therapy with intraaortic balloon pumping for treat-
coronary death: a comprehensive morphological classification ment of cardiogenic shock. Am Heart J 2006; 152:469 e1–e8.
scheme for atherosclerotic lesions. Arterioscler Thromb Vasc 79. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization
Biol 2000; 20:1262–1275. and long-term survival in cardiogenic shock complicating acute
61. Arbustini E, Dal Bello B, Morbini P, et al. Plaque erosion is a myocardial infarction. JAMA 2006; 295:2511–2515.
major substrate for coronary thrombosis in acute myocardial 80. Jacobs AK, Leopold JA, Bates E, et al. Cardiogenic shock caused
infarction. Heart 1999; 82:269–272. by right ventricular infarction: a report from the SHOCK registry.
62. Burke AP, Farb A, Malcom G, et al. Effect of menopause on J Am Coll Cardiol 2003; 41:1273–1279.
plaque morphologic characteristics in coronary atherosclerosis. 81. Joner M, Finn AV, Farb A, et al. Pathology of drug-eluting stents
Am Heart J 2001; 141:S58–S62. in humans: delayed healing and late thrombotic risk. J Am Coll
63. Stone GW, Ware JH, Bertrand ME, et al. Antithrombotic strat- Cardiol 2006; 48:193–202.
egies in patients with acute coronary syndromes undergoing 82. Kuchulakanti PK, Chu WW, Torguson R, et al. Correlates and
early invasive management: one-year results from the ACUITY long-term outcomes of angiographically proven stent thrombosis
trial. JAMA 2007; 298:2497–2506. with sirolimus- and paclitaxel-eluting stents. Circulation 2006;
64. Stone GW, Witzenbichler B, Guagliumi G, et al. Bivalirudin 113:1108–1113.
during primary PCI in acute myocardial infarction. N Engl J 83. Pendyala L, Jabara R, Hou D, et al. Review of percutaneous
Med 2008; 358:2218–2230. therapy for bifurcation lesions in the era of drug-eluting stents.
65. King SB III, Smith SC Jr., Hirshfeld JW Jr., et al. 2007 focused Minerva Cardioangiol 2008; 56:89–105.
update of the ACC/AHA/SCAI 2005 guideline update for 84. Lefevre T, Louvard Y, Morice MC, et al. Stenting of bifurcation
percutaneous coronary intervention: a report of the American lesions: classification, treatments, and results. Catheter Cardio-
College of Cardiology/American Heart Association Task Force vasc Interv 2000; 49:274–283.
on Practice guidelines. J Am Coll Cardiol 2008; 51:172–209. 85. Medina A, Suarez de Lezo J, Pan M. A new classification of
66. Grines C, Patel A, Zijlstra F, et al. Primary coronary angioplasty coronary bifurcation lesions. Rev Esp Cardiol 2006; 59:183.
compared with intravenous thrombolytic therapy for acute myo- 86. Pan M, Medina A, Suarez de Lezo J, et al. Follow-up patency of
cardial infarction: six-month follow up and analysis of individual side branches covered by intracoronary Palmaz-Schatz stent. Am
patient data from randomized trials. Am Heart J 2003; 145:47–57. Heart J 1995; 129:436–440.
67. Grzybowski M, Clements EA, Parsons L, et al. Mortality benefit 87. Poerner TC, Kralev S, Voelker W, et al. Natural history of small
of immediate revascularization of acute ST-segment elevation and medium-sized side branches after coronary stent implanta-
myocardial infarction in patients with contraindications to tion. Am Heart J 2002; 143:627–635.
thrombolytic therapy: a propensity analysis. JAMA 2003; 290: 88. Califf RM, Abdelmeguid AE, Kuntz RE, et al. Myonecrosis after
1891–1898. revascularization procedures. J Am Coll Cardiol 1998; 31:241–251.
68. Keeley EC, Boura JA, Grines CL. Primary angioplasty versus 89. Dahm JB, Ruppert J, Hartmann S, et al. Directional atherectomy
intravenous thrombolytic therapy for acute myocardial infarction: facilitates the interventional procedure and leads to a low rate of
a quantitative review of 23 randomised trials. Lancet 2003; 361: recurrent stenosis in left anterior descending and left circumflex
13–20. artery ostium stenoses: subgroup analysis of the FLEXI-CUT
69. Jneid H, Fonarow GC, Cannon CP, et al. Impact of time of study. Heart 2006; 92:1285–1289.
presentation on the care and outcomes of acute myocardial 90. Steigen TK, Maeng M, Wiseth R, et al. Randomized study on
infarction. Circulation 2008; 117:2502–2509. simple versus complex stenting of coronary artery bifurcation
70. Mehta RH, Harjai KJ, Cox DA, et al. Comparison of coronary lesions: the Nordic bifurcation study. Circulation 2006; 114:1955–
stenting versus conventional balloon angioplasty on five-year 1961.
mortality in patients with acute myocardial infarction undergo- 91. Teirstein PS. Kissing Palmaz-Schatz stents for coronary bifurca-
ing primary percutaneous coronary intervention. Am J Cardiol tion stenoses. Cathet Cardiovasc Diagn 1996; 37:307–310.
2005; 96:901–906. 92. Colombo A, Stankovic G, Orlic D, et al. Modified T-stenting
71. Laarman GJ, Suttorp MJ, Dirksen MT, et al. Paclitaxel-eluting technique with crushing for bifurcation lesions: immediate results
versus uncoated stents in primary percutaneous coronary inter- and 30-day outcome. Catheter Cardiovasc Interv 2003; 60:145–151.
vention. N Engl J Med 2006; 355:1105–1113. 93. Costa RA, Mintz GS, Carlier SG, et al. Bifurcation coronary
72. King SB III, Hannan EL. Mounting evidence for safety and lesions treated with the “crush” technique: an intravascular
improved outcomes of drug-eluting stenting: but is it the stent? ultrasound analysis. J Am Coll Cardiol 2005; 46:599–605.
Circulation 2008; 118:1783–1784. 94. Chevalier B, Glatt B, Royer T, et al. Placement of coronary stents
73. Mehran R, Lansky AJ, Witzenbichler B, et al. Bivalirudin in in bifurcation lesions by the “culotte” technique. Am J Cardiol
patients undergoing primary angioplasty for acute myocardial 1998; 82:943–949.
infarction (HORIZONS-AMI): 1-year results of a randomised 95. Kim YH, Park DW, Suh IW, et al. Long-term outcome of simul-
controlled trial. Lancet 2009; 374:1149–1159. taneous kissing stenting technique with sirolimus-eluting stent
74. Reynolds HR, Hochman JS. Cardiogenic shock: current concepts for large bifurcation coronary lesions. Catheter Cardiovasc Interv
and improving outcomes. Circulation 2008; 117:686–697. 2007; 70:840–846.
75. Chen JP. Of octopus pots and broken hearts. J Invasive Cardiol 96. Shaw LJ, Berman DS, Maron DJ, et al. Optimal medical therapy
2007; 19:403. with or without percutaneous coronary intervention to reduce
76. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization ischemic burden: results from the Clinical Outcomes Utilizing
in acute myocardial infarction complicated by cardiogenic shock. Revascularization and Aggressive Drug Evaluation (COURAGE)
SHOCK Investigators. Should we Emergently Revascularize trial nuclear substudy. Circulation 2008; 117:1283–1291.
Occluded Coronaries for Cardiogenic Shock. N Engl J Med 97. Hueb W, Lopes NH, Gersh BJ, et al. Five-year follow-up of the
1999; 341:625–634. Medicine, Angioplasty, or Surgery Study (MASS II): a random-
77. Chen EW, Canto JG, Parsons LS, et al. Relation between hospital ized controlled clinical trial of 3 therapeutic strategies for multi-
intra-aortic balloon counterpulsation volume and mortality in vessel coronary artery disease. Circulation 2007; 115:1082–1089.
acute myocardial infarction complicated by cardiogenic shock. 98. Nishigaki K, Yamazaki T, Kitabatake A, et al. Percutaneous
Circulation 2003; 108:951–957. coronary intervention plus medical therapy reduces the
[26/6/010/22:42:14] [388–403]
incidence of acute coronary syndrome more effectively than 101. Patel MR, Dehmer GJ, Hirshfeld JW, et al. ACCF/SCAI/STS/
initial medical therapy only among patients with low-risk coro- AATS/AHA/ASNC 2009 Appropriateness Criteria for Coronary
nary artery disease a randomized, comparative, multicenter Revascularization: a report by the American College of Cardiol-
study. JACC Cardiovasc Interv 2008; 1:469–479. ogy Foundation Appropriateness Criteria Task Force, Society for
99. Hlatky MA, Boothroyd DB, Bravata DM, et al. Coronary artery Cardiovascular Angiography and Interventions, Society of Tho-
bypass surgery compared with percutaneous coronary interven- racic Surgeons, American Association for Thoracic Surgery,
tions for multivessel disease: a collaborative analysis of individ- American Heart Association, and the American Society of
ual patient data from ten randomised trials. Lancet 2009; Nuclear Cardiology Endorsed by the American Society of Echo-
373:1190–1197. cardiography, the Heart Failure Society of America, and the
100. Serruys PW, Morice MC, Kappetein AP, et al. Percutaneous cor- Society of Cardiovascular Computed Tomography. J Am Coll
onary intervention versus coronary-artery bypass grafting for Cardiol 2009; 53:530–553.
severe coronary artery disease. N Engl J Med 2009; 360:961–972.
[8/7/010/14:7:29] [404–411]
31
Guiding catheters and wires

David W. M. Muller
GUIDING CATHETERS Catheters are constructed in a variety of shapes to cater

Introduction for differences in coronary anatomy. They all have a primary
Selection of an appropriate guiding catheter is fundamental to curve close to the tip and a secondary curve that varies
the success of every coronary interventional procedure. The according to the catheter size and shape. Subtle differences in
catheter should provide coaxial access to the coronary artery to construction along the length of the catheter give more support
facilitate passage of a guidewire, stability for delivery of bal- and torquability to the catheter shaft, and greater flexibility to
loons and bulky devices, and sufficient contrast flow to ade- segments closer to the tip (Fig. 31.2A, B). Variation in character-
quately visualize the coronary anatomy. It should have a soft istics of these zones or segments can greatly influence catheter
atraumatic tip to minimize the risk of injury to the coronary performance.
ostium. Changes in procedural complexity over the past two
decades have had important influences in guide catheter selec-
tion. Bulky inflexible devices such as the directional atherec-
Active Vs. Passive Guide Support
Depending on the design characteristics of the catheter, a guide
tomy catheter and early generation coronary stents required
may be best used passively or actively. Passive guide support
considerable support from the guide and gentle deflectable
relies on the inherent properties of the catheter and its interac-
curves distally. A progressive decrease in the bulk of devices,
tion with the walls of the aortic root. Passive support can be
an increase in their flexibility, and technical improvements in
increased by increasing the caliber of the guide (e.g., 8 Fr vs.
guiding catheter design have permitted a gradual reduction in
6 Fr), by selecting a shape that provides greater contact with the
guiding catheter caliber. Whereas only 59% of guiding catheters
contralateral aortic wall, or by selecting a catheter with a stiffer
were 6 Fr in 2000, the estimated distribution of catheter diam-
shaft and power zone. A guide that is used actively is manip-
eters in the United States market in 2007 was 79% 6 Fr, 14% 7 Fr,
ulated, for example, by deep seating into the proximal coronary
and 7% 8 Fr [Boston Scientific (BSC), Natick, Massachusetts,
artery (or beyond), to maximize back up support. This requires
U.S.]. The approximate inner diameter of contemporary 6 Fr
a soft atraumatic catheter tip and a flexible primary curve.
guiding catheters is 0.07 in., compared with 0.08 in. for 7 Fr
Some guides such as the Vista Brite Tip (Cordis Corporation,
guides and 0.09 in. for 8 Fr guides.
Miami Lakes, Florida, U.S.), Launcher (Medtronic, Santa Rosa,
California, U.S.), and Runway (Boston Scientific, Natick, Mas-
Guide Construction sachusetts, U.S.) are better suited to provide passive support,
Guiding catheters are typically constructed with three individ- whereas others such as the Mach1 (BSC), Zuma2 (Medtronic),
ual layers. The outer polyurethane or polyethylene layer pro- and Viking (Abbott Vascular, Redwood City, California, U.S.)
vides support, pushability, and curve retention with low catheters can be manipulated actively.
thrombogenicity (Fig. 31.1). The middle wire braid layer is a
flat or round, woven stainless steel or Kevlar layer that deter-
mines the catheter’s torque responsiveness and kink resistance. Catheter Shape and Variations
Tungsten may be incorporated to increase its radiopacity. The in Coronary Anatomy
weave of the braid varies (e.g., 2 2 or 2 4) and its density is The left main coronary and the right coronary artery usually
expressed as pic units. High pic values are associated with a arise horizontally from the left and right coronary cusps,
high-torque responsiveness of the catheter. The inner layer is respectively. It is not uncommon, however, for their position
usually PTFE (Teflon) to provide a low coefficient of friction to be anterior or posterior to the usual position and for their
and easy passage of devices through the catheter. Attached to takeoff to be superior or inferior to the horizontal (1) (Fig. 31.3).
the distal end is a soft radiopaque tip. Side holes may be The left main coronary artery may be long or short, or the left
incorporated close to the catheter tip, particularly in large anterior descending artery (LAD) and circumflex may arise
caliber catheters, to limit the hypoperfusion caused by partial from separate ostia (1,2). The right coronary artery often arises
occlusion of the coronary ostium. The benefit of the side holes high and anterior to its usual position. It may arise from the left
may, however, be offset by a reduction in visualization of the coronary cusp adjacent to the left main coronary artery. The
distal coronary artery, an increase in the volume of contrast circumflex can take origin from the proximal right coronary
used, and loss of pressure damping, an important indicator of artery (2). In this position, it can be difficult to cannulate using a
coronary flow limitation. guiding catheter that seats deeply in the proximal right
[8/7/010/14:7:29] [404–411]
GUIDING CATHETERS AND WIRES 405
Figure 31.1 Guide catheter construction. Source: Courtesy of

Boston Scientific.
coronary artery. Changes in the size and shape of the aortic root
and ascending aorta, and whether access is obtained from the
femoral or radial artery, also have an important influence on
the optimal size and shape of the guiding catheter.
Normal Anatomy
Traditional curves such as the left Judkins (JL or FL) catheter
provide adequate backup for many left coronary interventions
(Fig. 31.4). More supportive shapes such as the EBU, XB, Voda,
and DC curves provide better support for complex interven-
tions and, in some catheter laboratories, have become the
standard choice for all left coronary procedures. These shapes
have the added advantage of less angulated primary and
secondary curves that facilitates the passage of bulky devices. Figure 31.2 (A) Primary and secondary curves of Judkins left and
Gentle advancement with clockwise rotation can be used to right guiding catheters. (B) Functional zones of a right Judkins
selectively engage the LAD. Counterclockwise rotation can be guiding catheter. Source: Courtesy of Boston Scientific.
used to selectively engage the circumflex coronary artery. Other
curves that provide excellent backup in the left coronary artery
include the Amplatz curve (AL1 or AL2), the CLS (contralateral
support) curve, and the Q curve (Fig. 31.5A).
The right coronary artery is most commonly cannulated Radial Approach
using a right Judkins curve but, unless actively deep-seated, Judkins curves can be used for procedures performed using a
this provides relatively little support for complex anatomy. If left or right radial approach, but curves specifically designed
greater support is required, alternative curves include the for transradial interventions are often used. These include the
Hockey stick, Amplatz (AL1), XB or Voda right, allRight, and Kimny, MUTA, and Radial catheters, the RB catheter, and the
Kiesz Right (Fig. 31.5A). On occasions, the right coronary has a MAC catheter (Fig. 31.5B).
very superior take-off (Shepherd’s crook). If the artery is
calcified or severely diseased, instrumentation of these arteries Bypass Grafts
requires excellent backup using Shepherd’s crook, Amplatz Saphenous vein grafts to the right coronary artery are best
(AL1), and Hockey stick curves. In this, and other circum- approached with a multipurpose catheter if the takeoff is
stances in which the guide is deep seated, great care must be vertical, or a right Judkins or right coronary bypass (RCB)
taken to avoid guide catheter–related dissection of the proximal catheter if it is more horizontal. Vein grafts to the left coronary
artery. artery can be cannulated with a right Judkins, an IMA, a left
[8/7/010/14:7:29] [404–411]
Figure 31.3 Variations in orientation of the left and right coronary artery. The orientation of the left coronary (A) may be inferior (1),
orthogonal (2) or superior (3). In the AP orientation (B) it may lie anteriorly (1), orthogonally (2) or posteriorly (3). The right coronary artery may
also arise orthogonally (1), superiorly (Shepherd’s crook) (2), or from the inferior margin of the sinus (3). Source: From Ref. 1.
Registry in 1982 was only 59% (4). In the same year, Simpson
and colleagues reported the use of an over-the-wire system that
allowed a central guidewire to be advanced independently of
the dilatation catheter (5). This adjustment improved the suc-
cess and safety of the procedure by allowing the operator to
exchange or reshape the wire without removing the balloon
catheter, to minimize ischemia time by wiring the distal vessel
before attempting to cross with the balloon catheter, and to
exchange the balloon catheter without having to also remove
the guidewire. Soon thereafter, technical success rates appro-
aching 90% were reported (6).
Technical advances in guidewire technology followed. In
the mid 1980s, 0.018 and 0.016 in. wires were constructed from
a stainless steel core shaft that tapered distally and was covered
by a flexible stainless steel coil spring (Fig. 31.6A). The shaping
Figure 31.4 Judkins right and left curves. ribbon was added to the tip to improve shaping and steerability
of the wire. Visibility of the tip was enhanced by adding a
platinum alloy spring coil segment, and tip flexibility and
steerability were manipulated by varying the length and diam-
eter of the central core taper. Trackability was enhanced by
coating the distal coil with hydrophilic or hydrophobic materi-
coronary bypass (LCB), or an AL1 curve though none of these als (e.g., Teflon, silicone). This spring coil construction has
provides contralateral wall support. An IMA curve is used for remained dominant as the workhorse wire design in the two
interventions performed through an internal mammary artery decades since then, but other refinements have been introduced
graft. to improve wire performance in specific situations. In addition,
several manufacturers have developed wires with polymer
Anomalous Origins coatings covering part of or the entire distal segment to facil-
Right coronary arteries arising high and anteriorly, or low in itate passage of the wire through complex lesions and chronic
the right coronary cusp, can usually be cannulated with an total occlusions. The wire may or may not have an underlying
Amplatz left catheter. Circumflex arteries arising from the spring coil supporting the polymer.
proximal right coronary artery are best approached with a
short-tipped JR catheter, and AL1 or an AR1.
Wire Characteristics
GUIDEWIRES Contemporary wires can vary considerably in a variety of
Introduction important performance characteristics. The torque response of a
As originally conceived and developed by Gruntzig (3), balloon wire refers to the extent of wire tip rotation in response to
dilatation catheters had a closed end attached to a short, fixed, rotation of the wire shaft. Wires with high torque respond to
nonsteerable, and relatively atraumatic guidewire. The inability very fine movements of the wire shaft and are therefore more
to direct the wire to, and through, stenotic coronary lesions was steerable than less torquable wires. Wire support relates to the
a major limitation of the technique. As a consequence, the stiffness of the working length of the wire. Stiffer more sup-
primary technical success rate reported in the NHLBI PTCA portive wires more readily allow passage of a balloon catheter
[8/7/010/14:7:29] [404–411]
Figure 31.5 (A) Guide catheter shapes for transfemoral access. (B) Guide catheter shapes for transradial access. Source: Courtesy of
Boston Scientific.
or stent through noncompliant arteries. Wire trackability refers of arterial injury (dissection or perforation). Penetration power is
to the ease with which a wire can be advanced through a derived from the tip load and the area of the tip [tip load/tip
tortuous artery without buckling, kinking, or prolapsing. Push- area (kg/in2)].
ability relates to the extent to which pressure applied to the
shaft is transmitted to the wire tip. Durability and shape retention
describe the impact of repeated use of the wire on the integrity WORKHORSE WIRES
of the shaft and shape of the distal tip. Tip flexibility refers to the Coiled Tip Wires
ease with which a wire tip is deflected from an object. Very The most frequently used workhorse wires are still stainless steel
flexible, floppy-tipped wires are less likely to cause plaque coiled spring wires. These wires have a 0.014 in. stainless steel
disruption, dissection, or perforation than less flexible wires, shaft that tapers distally to a 30- to 40-cm long stainless steel core
but are also less likely to cross complex lesions and subtotal or (Table 31.2). The core typically tapers segmentally and termi-
total occlusions. Tip malleability refers to the ease with which the nates before the end of the overlying coils (Fig. 31.6A) with a
wire tip can be shaped. Tactile sense is the ease with which the shaping ribbon between the end of the core and the tip weld.
operator can recognize changes in movement of the wire tip. This wire construction provides good torque control, good
Tip load refers to the force required to buckle the distal tip pushability, and light support. Examples of this include the
(10 mm) of a wire. This is typically measured in grams. The Hi-Torque Floppy II wire (Abbott Vascular), the Asahi Light
higher the tip load, the greater the ease with which a wire will (Asahi Intec Co. Ltd., Nagoya-shi, Aichi, Japan), and the Boston
penetrate fibrous or calcified occlusions and the greater the risk Scientific Forte. The performance of this type of wire can be
[8/7/010/14:7:29] [404–411]
Figure 31.6 (A) Coiled tip workhorse wire with segmentally tapering stainless steel core. (B) Nitinol core improves durability of the wire
(increases resistance to kinking). The intermediate coils have been eliminated to improve trackability. (C) Upper image. Transitionless coil-to-
tip taper with a polymer sleeve and hydrophilic coating improves trackability (Whisper wire). Lower image. High tensile stainless steel core
with polymer sleeve covering all but the distal 5 mm designed to increase tactile sense. Core-to-tip design tapering to 0.009 in. to increase
penetration power (Hi-Torque Progress wire). (D) The Wiggle wire has a series of corrugations designed to exert a tangential force in heavily
calcified or previously stented arteries. Source: Courtesy of Abbott Vascular.
modified by increasing the diameter of the core to increase wire that are based on a unique method of processing the shaft and
support and torquability. This may require shortening or elim- core as a single piece with a transitionless, core-to-tip taper
ination of the intermediate coils between the shaft and the floppy (Tru-Torque) and eliminating the joint between the stainless
tip. Greater trackability and flexibility can be achieved by steel and the platinum coil segments at the tip.
reducing the diameter of the core, by reducing the abruptness
of the taper (transitionless core), by increasing the length of the Polymer-Coated Wires
taper to extend to the wire tip (core-to-tip), or by changing the To maximize trackability, wires have been developed with
shape of the taper (parabolic taper). polymer sleeves coating the distal tip (Fig. 31.6C). These
Other modifications to this original design and construc- include wires with polymer-coated coils (e.g., Hi-Torque
tion were introduced to further enhance wire performance. Whisper, Advance, Fielder, and Pilot wires), those with poly-
Changing the core material to nitinol (e.g., Hi-Torque Balance mer-coated core wires (e.g., PT2, PT Graphix, and ChoICE
Middleweight, IQ wire, Cougar), a highly elastic alloy of nickel PT wires), and hybrid combinations of both (BMW Universal,
and titanium, improved the durability (kink resistance) and Hi-Torque All Star) (Tables 31.1–31.4). The polymer is typically
flexibility of the wire while maintaining a moderate degree of impregnated with a material such as tungsten to add radio-
support (Fig. 31.6B). The price paid for this improvement may pacity to the tip. The reduced friction attributable to the poly-
be a reduction in torque control of the tip. Changing the core mer sleeve enhances passage of the guidewire through
material to a high tensile strength stainless steel (e.g., Hi- tortuous, severely diseased arterial segments, and through the
Torque Advance, Asahi Prowater/Rinato) also improved wire microchannels of chronic total coronary occlusions. The major
durability, shape retention, and torque control. Particularly disadvantages of the coating are a loss of tactile sense and an
good torque control is evident in the Asahi family of wires increased risk of arterial perforation. Wires without underlying
[8/7/010/14:7:29] [404–411]
Table 31.1 Wires Suitable for Standard Anatomy

Wire Manufacturer Core Tip Tip coating Tip stiffness Rail support
Hi-Torque Floppy Abbott Vascular SS Coil Silicone Floppy Light
BMW Abbott Vascular SE nitinol Coil Silicone Floppy Light
Light/Soft Asahi SS Coil Hybrid Floppy Light
Choice Floppy Boston SS Coil Hydrophilic Floppy Light
Forte Boston SS Coil Hydrophilic Floppy Light
Zinger Medtronic SS Coil Hydrophilic Floppy Light/Mod
Abbreviations: Boston, Boston Scientific Corporation; SS, stainless steel; SE, super elastic.
Table 31.2 Wires Suitable for Tortuous Arteries

IQ Boston LE nitinol Coil Silicone Floppy Light
Luge Boston SS PCC Hydrophilic Floppy Moderate
PT2 Boston LE nitinol Polymer Hydrophilic Intermediate Light/Mod
Prowater (Rinato) Asahi HTSS Coil Hydrophilic Floppy Moderate
Fielder Asahi SS PCC Hydrophilic Floppy Light
Whisper Abbott Vascular SS PCC Hydrophilic Intermediate Light/Mod
Advance Abbott Vascular HTSS PCC Hydrophilic Floppy Light
Wizdom Cordis SS PCC Silicone Floppy Light
Cougar Medtronic LE nitinol Coil Silicone Floppy Light/Mod
Abbreviations: Boston, Boston Scientific Corporation; LE, linear elastic; PCC, polymer-covered coils; SS, stainless steel; HTSS, high tensile
stainless steel.
Table 31.3 Wires Providing Extra Support

Extra S’Port Abbott Vascular SS Coil Silicone Floppy Extra
Stabilizer Plus Cordis Corporation SS Coil Silicone Floppy Extra
Thunder Medtronic SS Coil Silicone Floppy Extra
Ironman Boston SS Coil Silicone Floppy Super
Mailman Boston SS Coil Hydrophilic Floppy Super
Grand Slam Asahi SS Coil Silicone Floppy Super
Abbreviations: Boston, Boston Scientific Corporation; SS, stainless steel.
Table 31.4 Chronic Total Occlusion Wires

Miracle 3–12 Asahi SS Coil Silicone 3–12 g Moderate
Confianza/Conquest Asahi SS Coil Silicone 9–20 g Extra
Pilot 50–200 Abbott Vascular SS PCC Hydrophilic 2–5 g Moderate
Cross-it 100–400 XT Abbott Vascular SS Coil Hydrophilic 2–9 g Moderate
Progress 40–200T Abbott Vascular HTSS Hybrid Hydrophilic 5–13 g Moderate
Persuader Medtronic SS Coil Silicone 3–9 g Moderate
Shinobi Cordis SS PCC Hydrophilic 7–8 g Moderate
Crosswire Terumo Nitinol PCC Hydrophilic Moderate
Abbreviations: HTSS, high tensile stainless steel; LE, linear elastic; PCC, polymer-covered coils; SS, stainless steel.
tip coils may be difficult to shape and may have a reduced SPECIFIC PURPOSE WIRES
torque response. Some of these limitations have been addressed Angulated Lesions
with the Hi-Torque Progress family of wires (Abbott Vascular). Highly angulated lesions require wires that have exceptional
These have a polymer sleeve coating all but the distal 5 mm of torque control to allow fine movements of the tip. The tip
coils that are left bare to improve the tactile feel of the wire and should also be malleable with excellent shape retention.
to reduce the risk of perforation (Fig. 31.6C). Polymer-coated wires may provide an advantage in very
[8/7/010/14:7:29] [404–411]
diseased arterial segments but only if torque control and shape this subset of lesions. It is commonly necessary to use an array
retention are not compromised. Wires such as the Fielder and of wires to deal with the specific characteristics of individual
Fielder XT (Asahi) and the Whisper wire (Abbott Vascular) lesions. Most chronic occlusions that are more than six months
combine the benefits of excellent tip control with a polymer old have a proximal fibrous cap that is difficult to penetrate
coating and work well in this situation. with conventional wires. Several families of stiffer wires have
been developed with greater capacity to penetrate hard lesions
(Table 31.4). However, use of these wires in tortuous proximal
Angled Bifurcations/Tortuous Arteries
segments increases the risk of dissection or perforation of the
When the target vessel arises at a considerable angle from an
less diseased proximal artery. It is, therefore, recommended
adjacent artery (e.g., retroflexed circumflex, acutely angle OM
that softer wires are used to negotiate proximal bends. These
branch, or stented artery side branch), many wires prolapse
can then be exchanged using an over-the-wire balloon or
into the larger adjacent vessel rather than track around the
microcatheter for an appropriate total occlusion guidewire (7).
acute bend. This is particularly true of wires that have an
Wires that are specifically designed to penetrate the
abrupt transition zone or rapidly tapering core. Wires with a
proximal fibrous cap have a larger core diameter and short
transitionless core or long taper are most effective in negotiat-
core taper and are consequently stiffer and less flexible than
ing bends without prolapse of the wire tip. Suitable wires for
conventional wires (Table 31.4). Tip loads of these wires range
this situation include the Luge and IQ wires (BSC), the
from 3 to 20 g compared with 0.5 to 1 g for most soft wires.
Advance and Whisper wires (Abbott Vascular), and the Fielder
Some wires taper distally to a diameter of 0.009 to 0.010 in. to
wire (Asahi) (Table 31.2). The major downside to these wires is
further increase tip penetration power. Examples of the latter
a reduction in support for delivery of bulky devices.
include the Confianza (Conquest) Pro (Asahi), the Cross It XT
(Abbott Vascular), and Hi-Torque Progress (Abbott Vascular)
Calcified Tortuous Arteries wires. It is important to recognize that the stiffer the wire, the
When strong support is required to deliver a device to a distal greater the risk of subintimal dissection and perforation. Once
lesion in an artery that is calcified or very tortuous, extra the proximal fibrous cap has been penetrated with a stiff wire,
support wires are valuable. These typically have a large diam- more flexible wires, such as the Miracle 3 or a polymer-coated
eter inner core with a short taper at the tip. The PT2 Moderate wire, can be used particularly in long chronic total occlusions. It
Support (Boston Scientific), for example, has a core diameter of may then be necessary to exchange this wire for another stiff
0.097 in. compared with 0.075 in. in the PT2 Light Support. wire to penetrate the distal fibrous cap and enter the true lumen
Other examples of support wires are shown in Table 31.4 and of the distal vessel (7).
include the Balance Heavyweight (Abbott Vascular), the Stabi-
lizer (Cordis), and the ChoICE PT Extra Support wire (Boston
Retrograde Approach
Scientific). Even greater support can be obtained using the
An increasingly popular strategy for treating chronic total
Ironman (Boston Scientific), Grand Slam (Asahi), or Mailman
occlusions is the retrograde approach (8). This requires the
(Boston Scientific) wires (Table 31.3). The downside to these
operator to pass a wire through the contralateral artery into the
very rigid, heavily supportive wires is a tendency to distort
distal segment of the occluded vessel via septal or epicardial
angulated segments of the artery causing pleating and poten-
collaterals. Wires specifically designed for this approach, such
tially even transient closure of the artery. An alternative strat-
as the Fielder XT and Fielder FC, have excellent torque control
egy, if additional support is required, is to place a second wire
to facilitate negotiation of very fine and often tortuous (cork-
(buddy wire) alongside the first wire to help straighten curves
screw) collateral arteries. Most are also polymer coated to
in the artery and to increase the stability of the guiding catheter.
increase their trackability through these channels. Once the
distal vessel is entered, the wire can be exchanged for a stiffer,
Subtotal Occlusions/Short or Recent Total more penetrating wire using a microcatheter.
Occlusions
A major stimulus for the development of the plethora of wires
currently available was recognition that traditional wires do not OTHER SPECIALTY WIRES
adequately address the needs of operators treating chronic total Wiggle Wire
occlusions. Considerable effort has been expended in improv- This has a series of corrugations in the distal wire to promote
ing our understanding of the pathology of chronic occlusions. passage of the wire and balloon catheter through heavily
For example, the presence of microchannels and partial recan- calcified or previously stented lesions (Fig. 31.6D). The angu-
alization has implications for wire selection. For subtotal lated segment of the wire displaces it from the wall or the stent
chronic occlusions, and those with visible microchannels, poly- struts allowing free movement of the wire and balloon.
mer-coated wires with excellent tip control are very effective.
Occlusions due to recent myocardial infarction can usually be RotaWire
crossed readily with a soft or floppy tipped wire. Care does This dedicated rotational atherectomy wire has a 0.009 in. shaft
need to be taken to avoid dissection due to passage of the wire that tapers to 0.005 in. distally. Attached to the tip is a 2.2 or
into the base of an ulcerated or aneurysmal lesion. 2.8 cm long, 0.014 in. diameter spring coil tip. Floppy and Extra
Support versions of the wire are available.
Chronic Total Occlusions
Chronic total occlusions vary considerably in the duration of Distal Protection Wires
occlusion, in the extent of calcification, angulation, and proxi- Balloon dilatation and stenting of coronary or peripheral sten-
mal tortuosity, and in the presence of adjacent side branches. oses may liberate embolic material causing ischemic injury to
As a result, there is no single wire design that is best suited to the distal bed. Numerous guiding wires designed to collect and
[8/7/010/14:7:29] [404–411]
Figure 31.7 AngioGuard (upper panel) and FilterWire (lower two panels) distal protection devices. Source: Courtesy of Cordis Corporation
and Boston Scientific, respectively.
retrieve embolic debris are now available (Fig. 31.7) (9). The 2. Kimbiris D, Iskandrian AS, Segal BL, et al. Anomalous aortic
shaft of the wires is usually sufficiently robust to allow stenting origin of coronary arteries. Circulation. 1978; 58:606–615.
in most situations. If greater support is required, a buddy wire 3. Gruntzig AR. Transluminal dilatation of coronary artery stenoses.
Lancet 1978; 1:263.
system can be used.
4. Kent KM, Bentivolgio LG, Block PC, et al. Percutaneous trans-
luminal coronary angioplasty: report from the Registry of the
National Heart, Lung, and Blood Institute PTCA Registry. Am J
CONCLUSIONS Cardiol 1982; 49:2011–2020.
Although a vast and potentially daunting array of guiding 5. Simpson JB, Baim DS, Robert EW, et al. New catheter system for
catheter and guidewire designs is currently available, most coronary angioplasty. Am J Cardiology 1982; 49:1216–1222.
coronary interventions can be performed with a small number 6. Detre KM, Holubkov R, Kelsey SF, et al. Percutaneous trans-
of safe and reliable workhorse guides and wires. Specialty luminal coronary angioplasty in 1985–1968 and 1977–1981. The
wires and guides should be used less frequently to treat the NHLBI PTCA Registry. N Engl J Med 1988; 318:265–270.
specific lesions for which they were designed. 7. Ochiai M, Ashida K, Asaki H, et al. The latest wire technique for
chronic total occlusions. Ital Heart J 2005; 6:489–493.
8. Wu EB, Chan WW, Yu CM. Retrograde chronic total occlusion
REFERENCES intervention: tips and tricks. Catheter Cardiovasc Interv 2008;
1. Casserly IP, Franco I. Guides and wires in percutaneous coronary 72:806–814.
intervention. In: Ellis SG, Holmes DR, eds. Strategic Approaches 9. Moris C, Lozano I, Martin M, et al. Embolic protection in
in Coronary Intervention. 3rd ed. Philadelphia: Lippincott saphenous percutaneous interventions. EuroIntervention 2009; 5:
Williams and Wilkins, 2006:91–100. D45–D50.
[17/6/010/9:23:2] [412–428]
32
Coronary artery stenting

Stéphane Cook and Stephan Windecker
INTRODUCTION strength resulting in increased rates of restenosis and stent

The safety and efficacy of percutaneous coronary intervention thrombosis.
(PCI) has continuously improved since its inception more than
30 years ago. The breathtaking growth of PCI reflects its wide- Palmaz-Schatz Stent
spread acceptance as the preferred revascularization strategy In the late 1980s, Julio Palmaz, an Argentine radiologist,
compared with coronary artery bypass graft surgery (CABG). designed a vascular stent from a model taken from a piece of
Stents have remarkably improved the safety of PCI by reducing lathed metal. Together with Richard Schatz, a cardiologist from
periprocedural acute closure due to coronary dissection (1) and San Antonio, Texas, he modified the initial version of this
the need for emergent CABG (2). prototype into the first tubular slotted balloon-expandable
The basic principles underlying short-term efficacy are stent. In October 1987, the first peripheral Palmaz stent was
common to all coronary stents and include the following: implanted in Freiburg, Germany, and in December of the same
year the first Palmaz-Schatz coronary stent was implanted in
1. Increasing the arterial lumen by scaffolding the arterial
Sao Paolo, Brazil. The stents were crimped on the coronary
wall
angioplasty balloon by the interventional cardiologists, a
2. Trapping intimal flaps between the stent surface and
method which was prone to stent loss.
arterial wall
3. Sealing medial dissections Improvement of Bare-Metal Stents and Antiplatelet Regimen
The widespread acceptance of coronary artery stenting resulted
from the Belgian Netherlands STENT (BENESTENT) (5) and the
Historical Perspective Stent Restenosis Study (STRESS) (6) trials, which showed supe-
The word “stent” was coined in 1916 by Jan F. Esser, a Dutch riority of stents compared with balloon angioplasty in reduction
plastic surgeon, and referred to a dental impression compound of restenosis and need for repeat revascularization. Since then,
developed formerly by Charles Thomas Stent. The first vascular tremendous progress has been made in improving stent mate-
stent was developed and implanted in 1968 by Charles Dotter rial, design, and processing, resulting in superior deliverability
in a canine popliteal artery (3). and procedural success. The improved results with coronary
The first coronary stent resulted from discussions artery stenting over time were also related to expansion of the
between two Swedish expatriates in Switzerland: Hans Wall- indications for stent implantation and the discovery that dual
sten, a paper engineer, and Ake Senning, the chief cardiac antiplatelet therapy (instead of oral anticoagulation) lowered
surgeon collaborating with Andreas Grüntzig during the first both the incidence of stent thrombosis and hemorrhagic com-
coronary angioplasties in Zurich. The first coronary stent plications. On the basis of their efficacy, coronary artery stents
(Wallstent) was self-expanding and was developed by Medi- have emerged as the preferred tool of PCI and are currently
nvent in cooperation with Ulrich Sigwart. It was first deployed in more than 90% of procedures (7).
implanted in March 1986 by Jacques Puel (Toulouse, France)
in a 63-year-old male suffering from restenosis after balloon Drug-Eluting Stents
angioplasty of the left anterior descending (LAD) artery. The More recently, drug-eluting stents (DES) have been introduced.
first bailout stenting was performed by Ulrich Sigwart during DES deliver site-specific, controlled release of therapeutic
a live course (Lausanne, Switzerland) in June 1986 in a 50-year- agents. Heparin has been used as a stent coating in an attempt
old female suffering from occlusive dissection of the LAD to reduce the thrombogenic potential and risk of acute/sub-
artery after balloon angioplasty (Fig. 32.1). Shortly after these acute stent thrombosis (8). When used in the setting of acute
successful procedures, an unanticipated bane of stent implan- myocardial infarction (AMI), one study showed a reduced rate
tation emerged: stent thrombosis. To limit this serious compli- of stent thrombosis and recurrent myocardial infarction (MI) at
cation, aggressive anticoagulant regimens were introduced. 30 days with a heparin-coated stent (9). However, although
heparin has anti-inflammatory effects, no effect was observed
Gianturco-Roubin Stent on restenosis. Sirolimus-eluting stents (SES) were first
The Gianturco-Roubin stent, a balloon-expandable stent, had a implanted in 2001 and subsequently became the first DES
coil design manufactured from a single strand of stainless that significantly reduced the risk of restenosis inherent with
steel wire (4). The stent was approved in the United States in bare-metal stents (BMS). This was followed by paclitaxel-elut-
1993 for the treatment of coronary dissection during balloon ing stents (PES), which also consistently reduced the rate of
angioplasty. Similar to the Wallstent, the Gianturco-Roubin restenosis and the need for repeat revascularization procedures
stent had a great degree of flexibility but poor radial compared with BMS.
[17/6/010/9:23:2] [412–428]
CORONARY ARTERY STENTING 413
Figure 32.1 The first coronary stent implantations.
TECHNICAL CONSIDERATIONS Metallic Composition

Bare-Metal Stents Although nitinol, tantalum, cobalt chromium, and platinum
The key prerequisites for a modern coronary artery stent are alloys have all been tested in the coronary circulation, 316L
stainless steel remains the most frequently used component of
1. deliverability with favorable flexibility and low profile,
coronary stents. Cobalt chromium (L605 CoCr) alloy has become
2. radial strength to prevent elastic recoil (usually <4%) and
a more recent alternative. L605 CoCr is stronger, more radio-
limit foreshortening (usually <3%),
paque, and contains less nickel than 316L stainless steel. Stents
3. sufficient plaque coverage (usually 10–25%) to avoid tissue
manufactured from L605 CoCr feature thinner struts without
prolapse, and
compromising radial strength or radiographic visibility and
4. access to side branches with limited stent deformation when
provide improved deliverability compared with 316L stainless
opening struts for stent deployment in bifurcation lesions.
steel stents.
The available stents may vary in their metallic composi-
tion, strut design and thickness, delivery system, and coating.
These different parameters play an important role in deliver- Strut Design and Thickness
ability, visibility, scaffolding performance, and procedural suc- Basic stent characteristics—coil versus slotted tube versus mod-
cess. Some of the parameters can also influence the occurrence of ular, percent metal coverage, number of struts, strut thickness,
adverse events during the hospital stay (periprocedural myo- and strut morphology—are summarized in Figure 32.3. Cur-
cardial necrosis, stent thrombosis) and long-term follow-up rently, most stents have a slotted tube design, which can be
(restenosis) (10). further categorized into closed-cell and open-cell design. The
The importance of stent design on acute vascular injury closed-cell design provides better coverage of the luminal surface
and the subsequent proliferative response is well established. and conveys greater radial strength. Cell size is minimally
In animal models, changes in stent design lead to diverse affected with a closed-cell stent design deployed in a tortuous
degrees of vascular injury, thrombosis, and neointimal hyper- site. However, it is less flexible and may be more difficult to
plasia (11). Furthermore, stents that allow a circular rather than deliver in tortuous and calcified arteries. In addition, side
angular vessel lumen lessen neointimal proliferation (12). branch access may be more challenging. The open-cell design
Figure 32.2 illustrates the impact of stent design on restenosis allows for a greater flexibility of the stent and easier access to
and stent thrombosis. However, only a few randomized trials side branches. The drawbacks are a weaker radial strength,
have addressed the role of stent design on clinical outcome. changes in cell size in tortuous anatomy, and less coverage of
Compared with the Palmaz-Schatz stent, the Gianturco-Roubin the lumen, particularly on the outer curvature of the artery. To
II stent was shown to be inferior for the prevention of restenosis further improve flexibility, the number of crowns has been
(13). Several second-generation BMS have been directly com- increased accompanied by a decrease in strut length and strut
pared with the Palmaz-Schatz stent in noncomplex lesions thickness. Finally, the geometry of the cross-section of the strut
without showing differences in terms of stent thrombosis, has been improved and most of the struts are rounded to limit
restenosis, or major adverse cardiac events (MACE) (14–16). edge dissections and perforation.
[17/6/010/9:23:2] [412–428]
Figure 32.2 The effect of stent design on restenosis and early stent thrombosis.
Figure 32.3 Basic stent characteristics.
Delivery System Surface Coatings

Many balloon-expandable and a few self-expanding coronary Stent surface material importantly affects neointimal hyper-
stents have been developed for clinical use over the past several plasia. Various biological inert coatings (barrier coating), such as
years. None of the self-expanding stents has found broad carbon, gold, platinum, and phosphorylcholine, have been
application in the coronary circulation. developed to limit both restenosis and the risk of stent
[17/6/010/9:23:2] [412–428]
thrombosis. Gold was associated with an increase in restenosis elution, the geometric configuration of the platform is critical
and need for repeat revascularization (10). Experimentally, to accommodate the required dose of the agent on the drug-
polymeric gel paving of the arterial wall has been reported to carrying units (the struts) and to allow adequate diffusion to
render the injured arterial wall nonthrombogenic and to build a ensure optimal tissue drug levels. Strut-based drug delivery
diffusional barrier to circulating mitogenic factors, thus pre- has been shown to be highly spacing-dependent. Accordingly,
venting neointimal hyperplasia (17,18). Phosphorylcholine is a an increased strut number has been associated with higher
zwitterionic and highly hydrophilic molecule, which represents mean arterial wall drug concentrations, and inhomogeneous
the major phospholipid head group of the outer layer of normal strut placement has been shown to significantly affect local
cell membranes. Synthetic phosphorylcholine polymers can be concentrations (24).
physically adsorbed onto the stent surface and cross-linked to
the metal by g-irradiation. In vitro and ex vivo models have Coating
shown that phosphorylcholine coating inhibits adsorption of The stent coating consists of one to three layers (Table 32.1). The
proteins such as fibrinogen and albumin, and reduces platelet most important layer is the polymer, which contains the drug
adhesion and complement activation (19). In a porcine coronary and allows for drug elution into the arterial wall by contact
model, the implantation of phosphorylcholine-coated stents transfer. Supplemental layers are found in most DES and con-
showed equal degrees of inflammation and endothelialization sist of either top coatings to delay drug release (e.g., PBMA) or
as uncoated stents (20). Following encouraging results in terms base coatings to increase polymer adhesion to the stent struts
of neointimal hyperplasia reduction in the porcine model, a (e.g., Parylene C). Most polymers are durable (nonbioabsorb-
titanium-nitride-oxide-coated stent was tested against a stain- able), but newer generation stents use bioabsorbable polymer
less steel stent of otherwise identical design. Binary restenosis carriers. Coatings are typically spray coated or dip coated.
was reduced from 33% in the control group to 15% in the Polymers have been pivotal for the development of local
titanium-nitride-oxide-coated stent group (P ¼ 0.07) (21). drug delivery, and in particular DES. Polymeric materials act as
a drug reservoir and allow for controlled drug release over
Covered Stents time. The drug may be dissolved either in a reservoir sur-
The use of polytetrafluoroethylene-covered stents has been eval- rounded by a polymer film or within a polymeric matrix.
uated in saphenous vein graft interventions. The rationale is that a Controlled drug release can occur by diffusion, chemical reac-
covered stent may be able to entrap friable degenerated material, tion, or solvent activation. Biodegradable polymers allow drug
decrease the probability of distal embolization, and reduce neo- release by both drug diffusion and matrix degradation,
intimal hyperplasia. However, a randomized trial of 400 patients whereas nondegradable polymers enable drug release by par-
undergoing PCI of saphenous vein grafts yielded disappointing ticle dissolution (25). Early efforts to identify suitable polymers
results, showing no benefit in terms of restenosis or MACE over for stent coating were characterized by exuberant inflammatory
BMS (22). In the coronary circulation, the use of covered stents is and thrombotic responses resulting in excessive neointimal
confined to the emergency treatment of coronary perforations and hyperplasia and arterial occlusion (26). These adverse effects
to exclusion of coronary aneurysms (23). have been attributed in part to inappropriate polymer degra-
dation and the molecular weight of the compounds. More
recently, a wide variety of biocompatible polymers, some of
Drug-Eluting Stents which trigger no or minimal inflammatory response, have been
Apart from the delivery system and the platform, which are developed as carriers for DES. Furthermore, some stents have
basically the same as for a BMS, DES contain two specific parts: only an abluminal polymer coating (asymmetric coating)
the polymer coating and the drug. The four components of a (Fig. 32.5). In the investigation of new drugs for local delivery,
DES are summarized in Figure 32.4. With respect to drug it is therefore mandatory to address not only the drug itself but
also the biocompatibility of the polymeric carrier.
Drug
The drug aims to limit neointimal proliferation, and the drug’s
ideal profile should be characterized by a
1. wide therapeutic window,
2. low inflammatory potential,
3. selectivity for smooth muscle cell proliferation without
toxicity to the medial and adventitial cell layers, and
4. promotion of re-endothelialization.
The efficacy of candidate drugs is not only dependent on
biological activity in vitro but is also determined by local
pharmacokinetics and physicochemical drug properties. Drug
distribution is mediated by stent strut configuration and the
balance between convective and diffusive forces (24). Hydro-
philic drugs such as heparin readily permeate into tissue but
are also rapidly cleared. In contrast, lipophilic agents such as
paclitaxel or limus analogues are water-insoluble and bind to
hydrophobic sites in the arterial wall. Although both hydro-
Figure 32.4 Four components of a drug-eluting coronary stent. philic and hydrophobic drugs show large spatial concentration
gradients in the arterial wall, lipophilic agents distribute better
[17/6/010/9:23:2] [412–428]
Table 32.1 Basic Characteristics of Currently FDA-Approved Drug-Eluting Stents

Coating Elution
Trade name Platform Coating characteristics Drug duration
Cypher (Cordis) 316L stainless steel 12.6-mm thick 3-layer coating: Durable Sirolimus 30 days
BxVelocity stent Main coat: 10-mm PEVA, PBMA,
(140 mm struts) and drug
Top coat: 0.6-mm PBMA
Base coat: 2-mm Parylene C
Taxus (Boston Scientific) 316L stainless steel 16-mm thick single-layer coating: Durable Paclitaxel 90 days
Express2 stent Translute SIBS copolymer with
(132-mm struts) the drug
Endeavor-Sprint Cobalt Chrome Driver 5.3-mm thick 2-layer coating: Durable Zotarolimus 14 days
(Medtronic) stent (91-mm struts) Main coat: 4.3-mm
phosphorylcholine coating with
the drug
Base coat: 1-mm Parylene C
XIENCE V (Guidant, L605 Cobalt Chrome 7.8-mm thick fluropolymer Durable Everolimus 120 days
Abbott) and Promus ML Vision stent multilayer coating with drug
(Boston Scientific) (81-mm struts)
Figure 32.5 Drug-eluting stent coatings. Abbreviations: SES, sirolimus-eluting stents; PES, paclitaxel-eluting stents; EES, everolimus-
eluting stents; ZES, zotarolimus-eluting stents; BES, biolimus-A9-eluting stents.
and more homogenously into the arterial wall than hydrophilic inhibited neointimal proliferation (28). The kinetics of drug
drugs. To date, immunosuppressive (limus family) and anti- release from SES was investigated in vivo and showed that
proliferative (paclitaxel) drugs are used. tissue levels were maximal at 14 days and remain substantial
Most DES use drugs that are analogues of sirolimus (limus up to 28 days.
family). The principal therapeutic agents of the limus family The FDA-approved sirolimus-eluting Cypher stent (SES)
include sirolimus, tacrolimus, zotarolimus, everolimus, biolimus, (Cordis Corporation, Warren, New Jersey, U.S.) is based on the
and myolimus. These agents bind to the intracellular receptor BX Velocity stent (Cordis Corporation) made of stainless steel
FKBP-12 and inhibit a phosphoinositide 3-kinase mammalian with a strut thickness of 140 mm. Sirolimus is immersed in a 10
target of rapamycin, thereby reversibly inhibiting the growth to 15 mm thick durable polymer matrix (1:1 mixture of the
factor– and cytokine-stimulated cell proliferation in the G1 phase polymers polyethylenevinylacetate and polybutylmethacrylate)
of the cell cycle (27) (Fig. 32.6). Vascular smooth muscle cells are at a 30% drug-to-polymer weight ratio, with a sirolimus con-
usually quiescent, proliferate at low indices (<0.05%), and remain centration of approximately 140 mcg/cm2 stent surface area.
in the G0 phase of the cell cycle. However, stimulated by vascular Drug release occurs by passive diffusion. A slow release for-
injury or growth factors, vascular smooth muscle cells reenter the mulation has been derived by application of a diffusion barrier
cell cycle at G1 and advance into S phase. topcoat over the basecoat.
Sirolimus Sirolimus, a highly lipophilic drug, was the The principal studies evaluating SES were the first-in-
first member of the limus family to be used for prevention of man studies (29,30) (45 patients with focal lesions), the RAVEL
restenosis following PCI. Following experimental studies show- (31) trial (238 patients with single lesion, SES vs. BMS) and the
ing potent suppression of vascular smooth muscle cell prolif- SIRIUS trial (32) (1058 patients, SES vs. BMS). The sirolimus-
eration, local delivery of sirolimus from stents also effectively eluting Cypher stent received CE mark approval in Europe in
[17/6/010/9:23:2] [412–428]
Figure 32.6 Mechanism of action of SES versus PES stents. Abbreviations: SES, sirolimus-eluting stents; PES, paclitaxel-eluting stents.
April 2002 and was approved by the FDA in the United States reservoirs (Conor Medsystems, LLC, Menlo Park, California,
in May 2003. In a pooled analysis of individual patient data on U.S.) is currently being investigated as a novel DES platform.
4958 patients enrolled in 14 randomized trials (mean follow-up In vitro drug release mimics the elution curve of the sirolimus-
ranging between 1 and 5 years) comparing SES with BMS, there eluting Cypher stent with approximately 80% of drug released
was a significant 57% reduction in the combined risk of death, during the first month after stent implantation, and preclinical
MI, or reintervention in favor of the DES (33). Therapeutic studies indicate that coronary artery tissue levels of sirolimus
benefit was largely driven by a pronounced reduction in the closely match those achieved with the sirolimus-eluting Cypher
need for repeat revascularization (Fig. 32.7). stent.
Mural release of sirolimus from the bioresorbable poly- Everolimus Everolimus is a sirolimus derivative, whose
mer polylactide-co-glycolide embedded in large nondeforming binding activity to the FKBP-12 domain is threefold lower and
Figure 32.7 Kaplan–Meier curve showing a pro-

nounced reduction in the need for repeat revascu-
larization with SES versus BMS. Abbreviations:
SES, sirolimus-eluting stents; BMS, bare-metal
stents. Source: Adapted from Ref. 54.
[17/6/010/9:23:2] [412–428]
immunosuppressive activity in vitro is two- to fivefold lower 1548 patients. ZES did not achieve the prespecified secondary
than sirolimus. Nevertheless, the in vivo efficacy of everolimus endpoint, in-segment late loss (ZES 0.36 47 mm vs. PES 0.23
is as potent as sirolimus (34) and potentially associated with 0.45 mm; P ¼ 0.023). Similarly, in-stent late loss was significantly
less inflammation than SES or PES (35). higher with ZES (0.67 49 mm) than PES (0.42 50 mm; P
The FDA-approved everolimus-eluting XIENCE V stent < 0.001) with a nonsignificant trend in in-stent binary restenosis
(EES) is based on the Multilink Vision stent (Abbott Laborato- (ZES 13.3%; PES: 6.7%, P < 0.08). ZES achieved its primary
ries, Abbott Park, California, U.S.), a cobalt chromium stent endpoint of noninferiority regarding target vessel failure at
with a strut thickness of 81 mm. A combination of acrylic and nine months (ZES 6.6% vs. PES 7.2%; P ¼ 0.685). Although the
fluoro polymers allow for an everolimus concentration of rate of MI was lower at 30 days (ZES 0.8% vs. PES 2.3%; P ¼ 0.02),
approximately 100 mcg/cm2 stent surface in the absence of a largely related to fewer side branch occlusions, there were no
topcoat polymer. The stent is designed to release approximately significant differences in rates of death, cardiac death, or MI at 9
80% of the drug within 30 days after implantation. So far the and 12 months (43). Stent thrombosis occurred in 0.8% of ZES-
stent platform has been evaluated in the SPIRIT I, II and III treated and 0.1% of PES-treated patients.
trials. SPIRIT I (36) (60 patients) demonstrated the superiority The CE-approved zotarolimus-eluting Endeavor Resolute
of EES over the otherwise identical BMS with respect to in-stent stent (Medtronic Vascular, Santa Rosa, California, U.S.) is a
late loss (0.10 0.21 mm vs. 0.87 0.37 mm, P < 0.001) and modified stent platform, which uses the BioLinxTM polymer
repeat interventions (3.8% vs. 17.9%). In both SPIRIT II (37) and system for release of zotarolimus from the Driver stent surface.
III (38) (300 and 1002 patients, respectively), EES proved not BioLinx consists of three polymers, a hydrophilic C19 polymer (a
only noninferior but superior to PES regarding in-stent late loss mixture of hydrophobic hexyl methacrylate and hydrophilic
(SPIRIT I: 0.11 0.27 mm vs. 0.36 0.39 mm; P < 0.001, vinyl pyrrolidinone and vinyl acetate), water soluble polyvinyl
SPIRIT II: 0.16 0.41 mm vs. 0.30 0.53 mm; P ¼ 0.002). Binary pyrrolidinone, and a hydrophobic C10 polymer (hydrophobic
restenosis tended to be lower in EES- than PES-treated patients butyl methacrylate to provide adequate lipophilicity for zotar-
in each trial, a difference which became significant in a pooled olimus) and allows for a more delayed release of the same
analysis of both studies (in-stent restenosis EES 1.9%, PES 4.9%, zotarolimus concentration as on the original Endeavor Sprint
Psuperiority ¼ 0.02; in-segment restenosis EES 4.1%, PES 7.8%, stent (1.6 mcg/mm2 stent surface area) with approximately 50%
Psuperiority ¼ 0.039). Finally, MACE was significantly reduced in of drug released during the first 7 days, and 85% of drug released
favor of EES in SPIRIT III (4.6% vs. 8.1%; RR ¼ 0.56; 95% CI, at 60 days after stent implantation. In vitro assays showed low
0.34–0.94; P ¼ 0.03). Of note, the rate of MI tended to be lower levels of monocyte adhesion, inflammatory reaction, and tissue
with EES than PES at 30 days, presumably related to fewer side factor expression suggesting a high degree of biocompatibility
branch occlusions (EES 1.0% vs. PES 2.7%; HR ¼ 0.38; 95% CI, comparable to the original Endeavor platform (44).
0.14–1.02; P ¼ 0.06). The RESOLUTE FIM trial (45) (139 patients) showed an
Zotarolimus Zotarolimus is an equipotent analogue of in-stent late loss of 0.22 0.27 mm and in-segment binary
sirolimus in vitro and in vivo. Two stents eluting zotarolimus restenosis of 2.1% at nine months. Intravascular ultrasound
are currently available. Both stents are based on the Driver stent (IVUS) follow-up at nine months revealed a percent volume
platform (Medtronic, Inc., Minneapolis, Minnesota, U.S.) with a obstruction of 3.7 4.0% and late acquired stent malapposition
strut thickness of 91 mm, made of cobalt chromium alloy, and in 6.8% of cases. At 12 months, the rate of target lesion
have coatings with both hydrophilic and hydrophobic moieties. revascularization (TLR) was 0.8% and target vessel failure
The FDA-approved zotarolimus-eluting Endeavor stent 7.0%. This platform is currently being compared against the
(ZES) has a 5-mm thick phosphorylcholine layer with a zotar- EES in a large-scale clinical trial (RESOLUTE III).
olimus concentration of approximately 1.6 mcg/mm2 stent Biolimus-A9 Biolimus-A9 is a highly lipophilic, semi-
surface area. In contrast to the sirolimus-eluting Cypher stent, synthetic sirolimus analogue. The drug is immersed at a con-
which elutes approximately 80% of its drug during the first centration of 15.6 mcg/cm2 into a biodegradable polymer,
30 days, the Endeavor stent releases the same proportion of polylactic acid, which is applied solely to the abluminal surface
zotarolimus within 10 days only. This stent platform has been of a flexible stainless steel stent. On the basis of in vivo studies,
evaluated in the ENDEAVOR I (39), II (40), III (41), and IV (42) polylactic acid is fully converted to lactic acid at six months and
trials. the polymer is resorbed within nine months.
The ENDEAVOR I trial (39) (100 patients) was a non- The device utilized for the clinical evaluation of the
randomized, single-arm study and showed an in-stent late loss biolimus-A9-eluting stent (BES; Biosensors International
of 0.33 0.36 mm at four months and of 0.61 0.44 mm at Group, Ltd., Newport Beach, California, U.S.) is the S-stent,
12 months. In the ENDEAVOR II trial (40) (1197 patients), ZES which features a strut thickness of 112 mm and is made of laser-
proved superior to the otherwise identical BMS regarding the cut stainless steel (46). Whereas the BioMatrix I and II platforms
primary endpoint of target vessel failure (7.9% vs. 15.1%; P < share a parylene primer on the stent surface to allow for
0.001) as well as in-stent late loss (0.61 0.46 mm vs. 1.03 binding of the biodegradable polymer-drug layer to the stent
0.58 mm; P < 0.001) and in-segment binary restenosis (13.2% vs. surface, the more recent BioMatrix III platform is free of the
35%; p < 0.001). ENDEAVOR III (41) compared ZES with SES. parylene primer. Since the drug is co-released from the biode-
ZES was found inferior regarding late loss (in-stent: ZES 0.60 gradable polylactic acid polymer by means of hydrolysis, a
0.48 mm; SES: 0.15 0.34 mm, P < 0.001) and binary restenosis, lower proportion of biolimus-A9 (50%) is released at 30 days
but superior regarding late acquired stent malapposition (ZES compared with the sirolimus-eluting Cypher stent, followed by
0.5% vs. SES 5.9%; P ¼ 0.02) and rate of MI (SES 3.5% vs. ZES complete drug release at approximately 180 days.
0.6%; RR ¼ 0.18; 95% CI, 0.03–0.96; P ¼ 0.04). There were no The STEALTH I trial (120 patients, 2:1 randomization)
significant differences in rates of death, cardiac death, stent demonstrated the superiority of the biolimus-A9-eluting
thrombosis, repeat revascularization, MACE, or target vessel BioMatrix I stent over the otherwise identical BMS with respect
failure. ENDEAVOR IV (42) compared ZES with PES in to in-stent late loss (0.26 vs. 0.74; P < 0.001) and in-stent
[17/6/010/9:23:2] [412–428]
restenosis (3.9% vs. 7.7%, respectively; P ¼ NS) (47). BES was indefinitely. This paclitaxel-eluting Taxus stent platform gained
compared with SES in a large-scale randomized trial (LEAD- FDA approval in 2004. It has undergone extensive clinical
ERS) with a primary clinical endpoint in 1707 patients with investigation and proved highly effective at reducing the risk
2472 lesions when used in routine clinical practice. The primary of angiographic restenosis by approximately 75% and the risk
endpoint, a composite of cardiac death, MI, or clinically indi- of TLR by approximately 60% compared with BMS across
cated target vessel revascularization (TVR) within nine months, nearly all studied lesion and patient subsets (54).
occurred in 9.2% of patients treated with BES and 10.5% of In contrast to the TAXUS experience, nonpolymeric
patients treated with SES, thus establishing noninferiority release of paclitaxel from the V-Flex stent (Cook, Inc.,
(Pnoninferiority ¼ 0.003; RR ¼ 0.88; 95% CI, 0.64–1.19; Psuperiority Bloomington, Indiana, U.S.) failed to prove superiority over
¼ 0.39). Rates of cardiac death (1.6% vs. 2.5%; P ¼ 0.22), MI BMS. Paclitaxel embedded into the bioresorbable polylactide-
(5.7% vs. 4.6%; P ¼ 0.30), and clinically indicated TVR (4.4% vs. co-glycolide and eluted from large reservoirs (COSTAR stent
5.5%; P ¼ 0.29) were similar for both stent types. BES were also platform, Conor Medsystems) appeared promising in the
noninferior to SES in in-stent percent diameter stenosis (20.9% 10 mcg/30 day abluminal release formulation. However,
vs. 23.3%; Pnoninferiority ¼ 0.001; Psuperiority ¼ 0.26), the principal results of the randomized COSTAR II trial (1700 patients)
angiographic endpoint of the study (48). comparing the paclitaxel-eluting COSTAR stent with the pacli-
A BES using the NOBORI stent platform has been taxel-eluting TAXUS stent showed significantly higher in-stent
directly compared against PES in two studies, NOBORI I late loss (COSTAR 0.64 mm vs. TAXUS 0.26 mm; P < 0.0001)
phase 1 (85 patients, 2:1 randomization) and phase 2 and in-segment restenosis (18.7% vs. 6.7%; P ¼ 0.0002). At eight
(243 patients, 153 BES, 90 PES) (49,50). Both studies found months and prior to angiography, the clinical outcome with
BES to be noninferior and superior to PES regarding in-stent COSTAR was inferior to TAXUS for MACE (11.0 vs. 6.9%; P ¼
late loss (phase 1: 0.15 0.27 mm vs. 0.32 0.33 mm; Psuperiority 0.005) and clinically driven TVR (8.1% vs. 4.3%), leading to the
¼ 0.006), neointimal volume assessed by IVUS (phase 1: 3.8 withdrawal of the device for commercial use (55). Of note,
10.9 mm3 vs. 14.6 15.0 mm3; P ¼ 0.006) (49), and binary following burst release of 1 mcg of paclitaxel, only very little
restenosis (phase 2: BES 0.7% vs. PES 5.4%; P ¼ 0.049). When drug was released during the first 60 days, whereas the bulk of
compared with SES, the biolimus-eluting NOBORI stent paclitaxel (80%) was released thereafter. It has been suggested
(54 patients) was similar at nine months for the primary that drug release was insufficient during the early period after
endpoint of in-stent late loss (BES 0.10 0.26 mm vs. SES arterial injury at the time of maximal smooth muscle cell
0.13 0.44 mm; P ¼ 0.66) and in-segment restenosis (BES 3.3% replication, which underlines the importance of dose and tim-
mm vs. SES 6.3%; P ¼ 0.65) (50). Moreover, the same inves- ing of drug release for effective reduction of restenosis.
tigators assessed endothelium-dependent coronary vasomotion The principal studies evaluating the clinical use of PES
proximal and distal to the implanted stents using right atrial are TAXUS I (61 patients, simple lesions, PES vs. BMS) (56),
pacing at increasing heart rates. While SES showed significant TAXUS II (536 patients, simple lesions, slow- or moderate-release
vasoconstriction both proximal (2.3 10%) and distal (5.4 PES vs. BMS) (57), TAXUS IV (1314 patients, complex lesions, PES
9%) to the stented segment, BES showed more physiological vs. BMS) (58), TAXUS V (1156 patients, complex lesions, PES vs.
vasodilatation proximal (7.9 10%) and distal (6.1 8.0%) to BMS) (59), and TAXUS VI (448 patients, long lesions, PES vs.
the stented segment comparable with BMS (51). BMS) (60). In the pooled analysis of these randomized trials
Paclitaxel Paclitaxel stabilizes polymerized microtubules including 3513 patients, patients randomized to PES had signifi-
and enhances microtubule assembly, forming numerous unor- cantly lower rates of TLR (HR ¼ 0.46; P < 0.001) and TVR (HR ¼
ganized and decentralized microtubules inside the cytoplasm. As 0.62; P < 0.001), while no difference was observed with respect to
a result, cell replication is inhibited and this effect is seen death or MI compared with BMS (Fig. 32.8) (54).
predominantly in the G0/G1 and G2/M phases of the cell Two dedicated TAXUS studies focused on in-stent reste-
cycle. Paclitaxel was shown to effectively inhibit vascular smooth nosis following BMS implantation. Patients were randomly
muscle cell migration and proliferation (52). In addition, it has assigned to undergo angioplasty followed by brachytherapy
several favorable characteristics for stent-based local drug deliv- or PES implantation. The results of both TAXUS III (28 patients)
ery, such as a high degree of lipophilicity and a long-lasting (61) and TAXUS V–ISR trials (396 patients) (62) were favorable
antiproliferative effect following a single-dose application at low for PES. For instance, in TAXUS V–ISR, the cumulative MACE
concentrations. In the porcine restenosis model, implantation of rate was higher in patients treated with brachytherapy com-
stents dip-coated with paclitaxel at increasing doses resulted in a pared with PES (20.1% vs. 11.5%; P ¼ 0.02).
dose-dependent inhibition of neointimal formation at 28 days.
However, the beneficial effects of paclitaxel on neointimal for- Drugs and Stents Under Investigation
mation were complicated by local cytotoxic effects manifested as Several other antiproliferative drugs (i.e., novolimus, myolimus)
a decrease in medial wall thickness, focal neointimal and medial are under investigation, and aim to widen the therapeutic win-
wall hemorrhage, and cell necrosis (53). dow and increase the safety profile while lowering the dose.
The FDA-approved device currently used for stent-based Some investigators advocate combination therapy of two or more
paclitaxel delivery is the Express stent (Boston Scientific Cor- different drugs. Moreover, polymer-free DES platforms are
poration, Natick, Massachusetts, U.S.) made of stainless steel under evaluation. These stents (Translumina porous stent, Bio-
with a strut thickness of 132 mm. Paclitaxel is immersed into a matrix Freedom, MIV three core technology) use surface poros-
nonbiodegradable polymer matrix to control release of the drug ities rather than polymers as the drug reservoir.
(Fig. 32.6). Most studies have been performed with a slow- In parallel, various biodegradable stents are under scru-
release formulation, from which paclitaxel at a dose density of tiny. The concept of biodegradable stents as a temporary scaf-
1 mcg/mm2 is released with an initial burst phase over 2 days, folding and drug delivery unit during the vascular healing
followed by a slow release of paclitaxel for 10 days. Of note, process following stenting is particularly appealing and several
more than 90% of the drug remains sequestered in the polymer polymers have been tested for this purpose. An important
[17/6/010/9:23:2] [412–428]
Figure 32.8 Kaplan–Meier curve showing a significant reduction in the need for repeat revascularization with PES versus BMS.
Abbreviations: PES, paclitaxel-eluting stents; BMS, bare-metal stents. Source: Adapted from Ref. 54.
limitation in the past has been the polymer-mediated occur- activities (21% vs. 27%; P < 0.004), and better quality of life
rence of severe local inflammation (26). A magnesium alloy- (25% vs. 21%; P ¼ 0.003) (68). Secondly, restenosis is associated
based stent has undergone clinical evaluation, but results were with a low risk of MI. Thus, several studies reported a 2% to 19%
disappointing due to early recoil as well as restenosis. A more rate of restenosis-related MI (69–71). In addition, the treatment of
promising concept is the biodegradable BVS stent platform restenosis itself also is associated with a small but finite proce-
introduced by Abbott. It consists of a crystalline poly-L-lactide dural risk of mortality or MI (72). Finally, some observational
backbone on top of which everolimus is applied in a form of studies suggest a potential negative impact of restenosis on
poly-L-lactide acid. Although the initial version also showed prognosis. The outcome of 3363 patients treated with balloon
some recoil, a revised platform has been shown to provide angioplasty and repeat angiography was analyzed according to
satisfactory long-term patency in porcine coronary arteries with the presence (1570 patients) or absence (1793 patients) of reste-
little inflammation (63), and no early stent recoil or thrombosis nosis (73). Although survival at six years was similar in both
in the first human trial (64). groups (95% vs. 93%; P ¼ 0.16), patients with restenosis had an
Finally, an alternative to current DES is a stent surface increased rate of MI (15% vs. 12%; P ¼ 0.0001). A study of
modification that facilitates cell growth and limits neointimal 603 patients with diabetes who underwent balloon angioplasty
proliferation. One such platform is the Genous bioengineered observed a higher 10-year mortality in patients with restenosis
stent, with iridium oxide coating, EPC capture technology, nano- (24% without restenosis, 35% with nonocclusive restenosis, and
textured surfaces, or cell-specific peptide linkers (Affinergy, Inc., 59% with occlusive restenosis). Notably, occlusive restenosis
Durham, North Carolina, U.S.). was an independent predictor of mortality in multivariable
analysis (74). Another series of 2272 consecutive patients treated
with BMS between 1992 and 1996 reported a mortality rate of
BASIC CONSIDERATIONS
6.0% at four years in patients without and 8.8% in patients with
Benefits of Bare-Metal Stents
restenosis (P ¼ 0.02) (75). Multivariable analysis identified reste-
Routine implantation of BMS has virtually abolished the risk of
nosis and age as the only independent predictors of mortality.
abrupt closure and emergency CABG. Moreover, BMS moder-
These studies are limited because of their retrospective design,
ately lower the rates of restenosis (4–6,65–67).
small sample size, lack of predefined outcome variables, and
The STRESS (6) and the BENESTENT (5) trials compared
potential ascertainment bias. An association of restenosis with
elective stenting with angioplasty alone in native coronary
prognosis has not been observed in prospective randomized
arteries with short (<15 mm in length) de novo lesions and
clinical trials.
showed a 25% to 30% reduction in restenosis rate. Subsequent
stent trials confirmed reduced rates of restenosis and TLR.
Benefits of Drug-Eluting Stents
Clinical Importance of Restenosis Numerous randomized clinical trials and meta-analysis attest
The clinical consequences of restenosis depend on the degree of that DES reduce rates of restenosis and repeat TLR by 40% to
restenosis, area at risk subtended by the restenosed target lesion, 70% compared with BMS (31,59,76–78). In a collaborative net-
amount of recruitable collaterals, and lesion location. Restenosis work meta-analysis of 38 randomized trials involving 18,023
impacts clinical outcome in several ways. First, repeat revascula- patients, the rate of TLR was reduced by 70% with SES and by
rization procedures influence patients’ quality of life and well 58% with PES compared with BMS at four years (Fig. 32.9) (79).
being. In the Optimal Angioplasty versus Primary Stenting For the comparison of SES versus BMS, this benefit translated
(OPUS) I trial, a randomized comparison of routine and provi- into a need to treat seven patients to prevent one revasculari-
sional BMS implantation that prospectively assessed quality-of- zation event; for PES versus BMS the number needed to treat
life parameters, patients without restenosis had less frequent was eight patients. Although the greatest benefits of DES over
angina (17% vs. 22%; P ¼ 0.03), fewer limitations in physical BMS regarding TLR reduction are seen during the first year
[17/6/010/9:23:2] [412–428]
Figure 32.9 A collaborative network meta-analysis of randomized trials showed a reduction in cumulative incidence of target lesion
revascularization for drug-eluting stents (PES and SES) versus bare-metal stents. Abbreviations: PES, paclitaxel-eluting stents; SES,
sirolimus-eluting stents; BMS, bare-metal stents. Source: Adapted from Ref. 79.
after implantation, all long-term analyses indicate that the BMS (0.5–0.6%). The same holds true for late stent thrombosis: a
benefit is sustained at four years. meta-analysis of 5023 patients showed no difference in the
The data of randomized trials were corroborated in incidence of late stent thrombosis between DES and BMS
large-scale registries comparing BMS and DES without proto- (0.2% vs. 0.3%; P ¼ 1.00) (93). A network meta-analysis of
col-mandated angiographic follow-up. TVR amounted to 7.4% 18,023 patients also showed no difference in the rate of early
for DES and 10.7% for BMS (P < 0.0001) at two years in a and late stent thrombosis between SES, PES, and BMS (79),
propensity score matched study from Ontario (80), and to 4.0% although the average length of dual antiplatelet therapy was
and 7.5%, respectively (P < 0.001), at three-year follow-up in longer in patients treated with DES than BMS.
the one-stent cohort of the Swedish Coronary Angiography
and Angioplasty Registry (SCAAR) (81). In the Western Den-
Very Late Stent Thrombosis
mark Heart Registry, rates of TVR at 15 months were 4.6%
Conversely, case reports, observational studies, extended fol-
and 7.1%, respectively (adjusted RR ¼ 0.57; 95% CI, 0.48–0.67;
low-up of trials comparing DES with BMS, and meta-analysis
P < 0.001) (82).
of randomized trials indicate that very late stent thrombosis is
more common with DES than BMS. Pooled analysis of four
Risks of Drug-Eluting Stents randomized trials (1748 patients) comparing SES and BMS and
DES delay healing and impair endothelialization as docu- of five randomized trials (3513 patients) comparing PES with
mented in necropsy studies (83,84) and clinical investigations BMS revealed significantly higher rates of very late stent
(85–87). IVUS studies show a higher incidence of incomplete thrombosis with DES (0.6% SES vs. 0% BMS, P ¼ 0.03; 0.7%
stent apposition with DES than BMS, and evidence of vessel PES vs. 0.2% BMS, P ¼ 0.03) (54). Similarly, a systematic review
remodeling following DES implantation is evident in patients of 14 trials comparing SES with BMS revealed that very late
with very late stent thrombosis (88–90). In addition, the drugs stent thrombosis was more frequent with SES (0.3% vs. 0.04%;
released from the polymer might have thrombogenic effects P ¼ 0.02) (33).
themselves; paclitaxel and sirolimus enhance endothelial tissue It should be noted, however, that these analyses were
factor expression, the principal activator of the coagulation based on primary stent thrombosis events (without intercurrent
cascade that activates factors IX and X (91,92). DES could revascularization). Secondary stent thrombosis events follow-
therefore predispose to thrombotic stent occlusion. ing intercurrent revascularization procedures were censored
and, therefore, could have disadvantaged devices with low
Early and Late Stent Thrombosis reintervention rates (DES) compared with high reintervention
Definitions of early, late, and very late stent thrombosis are rates (BMS). One study reanalyzed stent thrombosis events
summarized in Figure 32.10. A meta-analysis of six studies according to each of the newly proposed Academic Research
comparing SES with BMS in 2963 patients and of five trials Consortium (ARC) categories. The overall incidence of stent
comparing PES with BMS in 3513 patients demonstrated sim- thrombosis did not differ between DES and BMS, and differ-
ilar rates of early stent thrombosis for both types of DES and ences in the incidence of very late stent thrombosis diminished
[17/6/010/9:23:2] [412–428]
Figure 32.10 Definitions of early, late, and very

late stent thrombosis. Abbreviation: PCI, percuta-
neous coronary intervention.
resulting from more events following intercurrent revasculari- INDICATIONS

zation in the BMS group (94). Bare-Metal Stents
Treatment of Abrupt or Threatened Closure
After Balloon Angioplasty
Net Safety Outcome of Drug-Eluting Stents Coronary stents effectively scaffold coronary dissections com-
In balancing the risks and benefits of DES compared with BMS
plicating balloon angioplasty, and the availability of stents has
from a patient’s perspective, the assessment of safety endpoints
nearly eliminated the need for emergency CABG (<1% of all
such as overall mortality and MI are most important. The safety
PCI procedures). The initial approval of coronary stents for this
of DES compared with BMS has been analyzed in several
“bailout” indication was based on a multicenter registry of
observational studies and meta-analysis of randomized con-
patients with angioplasty complications who were treated with
trolled trials. It must be noted, however, that observational
the Gianturco-Roubin II stent (4).
studies cannot reliably determine small-to-moderate risks or
benefits of a therapeutic intervention (95). Factors associated Reduction of Restenosis and the Need
with the selection of stent type and confounded by indication of Repeat Revascularization Procedures
are uncontrollable. In light of these limitations, the results of Randomized clinical trials have shown that coronary artery
registries do not provide a definitive answer and should be stent implantation in de novo lesions of native coronary arteries
interpreted with caution. with a diameter >3.0 mm improves short- and long-term out-
Stone and colleagues observed no difference between comes compared with balloon angioplasty. The principal stud-
PES and BMS regarding death (6.1% vs. 6.6%; P ¼ 0.68) and ies in support of these findings are the STRESS (6) (410 patients;
MI (7.0% vs. 6.3%; P ¼ 0.66) at four years in a patient-level restenosis, stent vs. balloon angioplasty: 32% vs. 42%; P ¼
meta-analysis of five trials comprising 3513 patients (54). Sim- 0.046), BENESTENT (5) (520 patients; 22% vs. 33%; P ¼ 0.02),
ilarly, Kastrati and colleagues observed similar rates of death BENESTENT II (8) (827 patients; heparin-eluting stent vs. bal-
(6.0% vs. 5.9%) and death or MI (9.7% vs. 10.2%) for SES and loon angioplasty: 16% vs. 31%; P ¼ 0.0008), LAD stenting
BMS in a meta-analysis of 14 trials with 4958 patients at five- versus angioplasty (96) (120 patients with isolated stenosis of
year follow-up (33). In addition, a collaborative network meta- the proximal LAD; 19 vs. 40%; P ¼ 0.02), and START (97)
analysis of 38 randomized trials involving 18,023 patients (452 patients; 22% vs. 37%; P < 0.002) trials. The REST (98) trial
comparing first-generation DES with BMS showed similar was a randomized study of 383 patients with prior restenosis
rates of death and cardiac death at four-year follow-up (79). after angioplasty who were assigned to Palmaz-Schatz stent
The apparent paradox between increased rates of very late stent implantation or repeat angioplasty. Angiographic restenosis
thrombosis and lack of increased rates of mortality and MI with was lower among stent-treated patients (18% vs. 32%).
first-generation DES has been explained by the very low inci-
dence of stent thrombosis, the offsetting effect of reduced rates Stenting in Selected Subsets of Patients/Procedures
of restenosis and increased rates of stent thrombosis, and the Saphenous vein grafts Two studies [SAVED (99),
potential of secondary stent thrombosis following intercurrent 220 patients; VENESTENT (100), 150 patients] randomly assigned
revascularization for BMS restenosis. patients with de novo stenosis in saphenous vein grafts to
[17/6/010/9:23:2] [412–428]
treatment with Palmaz-Schatz stent implantation or balloon restenosis (8.3% vs. 25.5%; RR ¼ 0.33; 95% CI, 0.19–0.56; P <
angioplasty. In both studies, restenosis was not different between 0.001) and TVR (7.2% vs. 18.8%; RR ¼ 0.38; 95% CI, 0.22–0.66; P <
patients treated with BMS and balloon angioplasty (SAVED: 37% 0.001) with SES (52). The benefit was, however, largely restricted
vs. 46%; VENESTENT: 19% vs. 36%), while the incidence of to patients with small-vessel disease (reference vessel diameter
MACE was lower at 6 and 12 months in the BMS group. <2.8 mm; restenosis 7.0% SES vs. 34.2% BMS; P < 0.001). In
Total coronary occlusions Three studies [SICCO (101), 119 contrast, binary restenosis was similar in arteries with a reference
patients; TOSCA (102), 410 patients; GISSOC (103), 110 patients] diameter of 2.8 mm or more (mean RVD 3.1 mm; 10% SES vs. 13%
compared the outcome of patients treated with angioplasty alone BMS; P ¼ 0.52) (108). Another randomized trial comparing newer
or implantation of a Palmaz-Schatz stent (TOSCA: heparin-elut- generation DES with BMS in 826 patients without routine angio-
ing). In the SICCO study, 119 patients with successful recanaliza- graphic follow-up observed lower rates of TVR at 18 months
tion followed by angioplasty of a chronic total occlusion were (7.5% vs. 11.6%; P ¼ 0.05). A stratified subgroup analysis of this
assigned to no further intervention or to Palmaz-Schatz stent trial indicated that the benefit was limited to patients who
placement. Stent implantation was associated with a 30% to 50% received at least one small stent (<3.0 mm diameter; TVR HR
lower incidence of restenosis in all studies. 0.44 in favor of DES; P ¼ 0.02), whereas no difference was
ST-elevation myocardial infarction The CADILLAC observed in patients who received only large stents (3 mm)
(104) trial evaluated the use of abciximab and stents in a 2 2 (109). The Ontario propensity score matched comparison of DES
factorial design in 2082 patients with ST-elevation myocardial and BMS showed that DES were particularly beneficial in
infarction (STEMI). At six months, angiographic restenosis patients with small vessels (<3 mm) and long lesions
occurred in 41% treated with angioplasty compared with 22% (20 mm) (80). The number needed to treat in patients with
of those undergoing stent implantation (P < 0.001). The PAMI small vessels ranged from 0 to 12 in patients with diabetes to 27 to
stent (105) trial randomly assigned 900 STEMI patients to treat- 83 in nondiabetic patients. Similarly, the number needed to treat
ment with balloon angioplasty or implantation of a Palmaz- in patients with long lesions ranged from 10 to 23 in patients with
Schatz heparin-coated stent. After six months, angiographic diabetes, to 27 to 53 in nondiabetic patients. Accordingly, DES
restenosis was less frequent among patients treated with BMS seem particularly useful in arteries with a reference diameter
than balloon angioplasty (20% vs. 33%; P < 0.001). In the 3 mm and long lesions, whereas BMS remain a valuable alter-
FRESCO (106) trial, 150 STEMI patients who underwent suc- native for larger arteries with discrete lesions.
cessful balloon angioplasty were randomly assigned to stent
implantation or no further intervention. At six months, the
incidence of restenosis or reocclusion was 17% in the stent CLINICAL ASPECTS
group compared with 43% in the angioplasty group (P ¼ 0.001). Procedure-Related Factors
Better angiographic results correlate with improved clinical
outcome. Uncovered dissections and intramural hematoma
Drug-Eluting Stents are associated with abrupt closure. Stent underexpansion is
The therapeutic benefit of DES is largely related to the powerful associated with both restenosis and stent thrombosis after BMS
inhibition of neointimal hyperplasia, which is particularly and DES implantation. Therefore, an adequate minimal luminal
important in patients suffering from diabetes mellitus or in diameter following stent implantation should be reached.
smaller vessels less able to accommodate neointimal tissue. Along with improvements in stent design, observations made
by IVUS led to changes in deployment strategy, emphasizing
Patients with Diabetes the importance of complete apposition of stent struts to the
PCI in patients with diabetes is associated with an increased risk arterial wall. However, routine use of IVUS to guide stent
of restenosis, repeat revascularization procedures, stent throm- implantation has not been shown to improve clinical outcomes.
bosis, and mortality. Although a recent meta-analysis of four A minimal stent area of >5 mm2 is usually considered satis-
trials suggested that SES conferred an increased risk of death in factory. Avoidance of long stented segments and stent overlap
patients with diabetes compared with BMS (107), this finding was
should be considered. For bifurcation lesions, the strategy of
not confirmed in a larger analysis of 14 trials (33). Results from a provisional side branch stenting is preferred, and techniques
30-trial network meta-analysis of 3762 patients with diabetes using two stents with excessive stent overlap like the “crush” or
failed to show any difference regarding the combined endpoint
“culotte” techniques should be avoided (110,111).
of death or MI between DES and BMS (79). In an updated analysis
focusing on diabetic patients, the risk of TLR was reduced by 62%
with PES and 71% with SES compared with BMS, a benefit Antiplatelet Therapy
similar to that observed with nondiabetic patients. In 3751 pairs An important prerequisite for the widespread use of coronary
of patients treated with either DES or BMS in the Ontario study, artery stents was the modification of the antithrombotic regi-
TVR was significantly reduced with DES in all subsets of indi- men, replacing the combination of aspirin and warfarin with
viduals with diabetes, with the exception of those with large dual antiplatelet therapy consisting of aspirin and a thienopyr-
vessels (>3 mm) and short lesions (<20 mm) (80). As a result of idine. In conjunction with improved deployment techniques,
the higher baseline risk of restenosis, the absolute reduction in dual antiplatelet therapy has led to a significant reduction of
repeat revascularization is more pronounced in diabetic than acute and subacute stent thrombosis from 3% to 5% to currently
nondiabetic populations, and DES should be recommended in <1% in elective patients. Most of the BMS studies used a two-
this patient subgroup. to four-week regimen of thienopyridine therapy. However,
prolonged use of thienopyridine (9–12 months) therapy had
Small Vessels, Long Lesions been shown beneficial in patients undergoing PCI with BMS
A small, randomized trial (500 patients) comparing SES with (CREDO study) (112) as well as in patients with acute coronary
thin-strut BMS showed significantly lower angiographic syndromes.
[17/6/010/9:23:2] [412–428]
In patients treated with DES, most studies with SES used with unexplained thrombocytopenia or established allergy to
a two- to three-month regimen, whereas studies with PES thienopyridines, and all those in whom compliance with
prescribed a six-month regimen of thienopyridine therapy. extended dual antiplatelet therapy cannot be ensured.
However, the optimal duration of dual antiplatelet therapy
following DES implantation is not well established. The
BASKET-LATE study observed more ischemic adverse events
Lesion-Related Restrictions
Severe calcification preventing sufficient stent expansion should
in DES-treated than in BMS-treated patients during the obser-
be considered as a relative contraindication to stenting. Several
vation period between months 7 and 18, when clopidogrel
studies have identified acute coronary syndrome at the time of
treatment had been discontinued (113). The impact of thieno-
DES implantation as a predictor of stent thrombosis (116,120).
pyridine treatment on long-term outcome following DES
Potential explanations for this observation include the large
implantation has been examined in a single-center observatio-
thrombus burden encountered during PCI for STEMI, which
nal study of 4666 patients (3165 BMS patients, 1501 DES
seems to predispose to early and late stent thrombosis (121).
patients) (114). Using landmark analyses of patients event-
IVUS studies revealed a high incidence of incomplete stent
free at 6 and 12 months, the investigators assessed the risk of
apposition in patients with very late stent thrombosis, which
death and MI at two years stratified according to stent type and
could be related to thrombus resolution or positive arterial
self-reported thienopyridine intake. Although thienopyridine
remodeling in patients with STEMI (88). Although a recent
use inferred no advantage in patients treated with BMS who
meta-analysis found DES to reduce TVR without excess risk
were event-free at six months, it was a significant predictor of
when implanted in the setting of STEMI compared with BMS
lower adjusted mortality (2.0% with thienopyridine therapy vs.
(122), the clinical outcome of DES in this setting requires further
5.3% without; P ¼ 0.03), and death or MI (3.1% vs. 7.2%,
study.
respectively; P ¼ 0.02) in patients treated with DES. More
recently, Airoldi and colleagues investigated the impact of
thienopyridine discontinuation on the risk of stent thrombosis SPECIAL ISSUES
following DES implantation at various time points (115). Dis- Noncardiac Surgery After Drug-Eluting
continuation of the thienopyridine within the first six months of Stent Implantation
DES implantation conferred an increased risk of stent throm- Surgery poses a risk to patients with coronary artery disease,
bosis (HR ¼ 13.7; 95% CI, 4.0–47; P < 0.001); however, beyond particularly those after stent implantation due to antiplatelet
six months the increased risk was no longer apparent (HR ¼ therapy withdrawal, increased platelet aggregation (123), and
0.94; 95% CI, 0.30–3.0; P ¼ 0.92). Of note, 23% of patients with decreased fibrinolysis during the perioperative period. A recent
late stent thrombosis in a large, two-center cohort were on dual study of 103 stent patients undergoing noncardiac surgery
antiplatelet therapy, indicating that prolonged therapy may not found a 5% mortality and 45% complication rate (124). It is,
be a panacea for late stent thrombosis (116). therefore, essential to determine whether the procedure can be
Newer antiplatelet agents such as the thienopyridine postponed by 1 to 3 months after BMS and 12 months after DES
prasugrel might have a beneficial effect in further reducing implantation, and whether dual antiplatelet therapy can be
early stent thrombosis by means of a more rapid and consistent maintained throughout the perioperative period. If the risk of
inhibition of platelet aggregation. The incidence of stent throm- bleeding is judged unacceptably high, clopidogrel should be
bosis was more than halved in the TRITON-TIMI 38 study discontinued no longer than 5 days before the procedure and
comparing prasugrel and clopidogrel, with most of the benefit resumed within 48 hours; aspirin (80–100 mg daily) should not
emerging during the early period after stent implantation (stent be interrupted (125). Whether the addition of heparin (unfrac-
thrombosis with prasugrel 1.1% vs. clopidogrel 2.4%; HR ¼ tionated or low molecular weight) or glycoprotein IIb/IIIa
0.48; 95% CI, 0.36–0.64; P < 0.001) (117). With the currently inhibitors is useful for prevention of ischemic events in the
available evidence, it seems reasonable to maintain dual anti- perioperative period is not known.
platelet therapy for a duration of 6 to 12 months or even
indefinitely in patients treated with first-generation DES if
they are at low risk of bleeding. The adherence to a 12-month CONCLUSIONS
regimen of dual antiplatelet therapy after DES implantation has The introduction of coronary stenting successfully addressed
been endorsed by a recent science advisory report (118) as well the risk of abrupt closure after PCI. Moreover, systematic stent
as by a clinical alert issued by the Society of Cardiac Angiog- implantation has importantly reduced the risk of restenosis and
raphy and Interventions DES Task Force (119). therefore the need of repeat revascularizations compared with
balloon angioplasty. More recently, the introduction of DES has
further decreased the risk of restenosis. This benefit appears
LIMITATIONS AND CONTRAINDICATIONS particularly pronounced in patients at increased risk of reste-
Patient-Related Restrictions nosis such as diabetic patients and those with small vessels and
The indication for stent implantation versus an alternative treat- long lesions. While early and late stent thrombosis occur with
ment should be carefully weighed against the risk of bleeding similar frequency with DES and BMS, very late stent thrombo-
and thrombosis in any patient. Of special interest are patients at sis is more frequently encountered following DES implantation,
increased risk of bleeding, those scheduled for elective major potentially related to delayed healing and re-endothelialization.
surgery, patients with gastrointestinal disorders preventing full Notwithstanding, long-term follow-up to four years indicates
absorption of thienopyridines, and those requiring oral antico- no difference between DES and BMS regarding overall mortal-
agulation due to atrial fibrillation, pulmonary embolism, left ity, and the combined endpoint of death or MI. The increased
ventricular aneurysms, or prosthetic heart valves. In these risk of very late stent thrombosis with DES may be too small to
patients, when a coronary stent is mandatory, BMS should be have a measurable impact on clinical outcome, may be offset
preferred. Furthermore, DES are contraindicated in individuals partly by the reduction in repeat revascularizations, and be
[17/6/010/9:23:2] [412–428]
balanced by secondary stent thrombosis events following inter- balloon injury in the rat: the roles of medial and luminal factors in
current revascularization of BMS. arterial healing. Proc Natl Acad Sci U S A 1996; 93:13188–13193.
19. Yu J, Lamba NM, Courtney JM, et al. Polymeric biomaterials:
influence of phosphorylcholine polar groups on protein adsorption
REFERENCES and complement activation. Int J Artif Organs 1994; 17:499–504.
1. Sigwart U, Puel J, Mirkovitch V, et al. Intravascular stents to 20. Whelan DM, van der Giessen WJ, Krabbendam SC, et al. Bio-
prevent occlusion and restenosis after transluminal angioplasty. compatibility of phosphorylcholine coated stents in normal por-
N Engl J Med 1987; 316:701–706. cine coronary arteries. Heart 2000; 83:338–345.
2. Lindsay J, Hong MK, Pinnow EE, et al. Effects of endoluminal 21. Windecker S, Simon R, Lins M, et al. Randomized comparison of
coronary stents on the frequency of coronary artery bypass a titanium-nitride-oxide-coated stent with a stainless steel stent
grafting after unsuccessful percutaneous transluminal coronary for coronary revascularization: the TiNOX trial. Circulation 2005;
vascularization. Am J Cardiol 1996; 77:647–649. 111:2617–2622.
3. Dotter CT. Transluminally-placed coilspring endarterial tube 22. Turco MA, Buchbinder M, Popma JJ, et al. Pivotal, randomized
grafts. Long-term patency in canine popliteal artery. Invest U.S. study of the Symbiottrade mark covered stent system in
Radiol 1969; 4:329–332. patients with saphenous vein graft disease: eight-month angio-
4. Roubin GS, Cannon AD, Agrawal SK, et al. Intracoronary stent- graphic and clinical results from the Symbiot III trial. Catheter
ing for acute and threatened closure complicating percutaneous Cardiovasc Interv 2006; 68:379–388.
transluminal coronary angioplasty. Circulation 1992; 85:916–927. 23. Eshtehardi P, Cook S, Moarof I, et al. Giant coronary artery
5. Serruys PW, de Jaegere P, Kiemeneij F, et al. A comparison of aneurysm: imaging findings before and after treatment with a
balloon-expandable-stent implantation with balloon angioplasty polytetrafluoroethylene-covered stent. Circ Cardiovasc Interv
in patients with coronary artery disease. Benestent Study Group. 2008; 1:85–86.
N Engl J Med 1994; 331:489–495. 24. Hwang CW, Wu D, Edelman ER. Physiological transport forces
6. Fischman DL, Leon MB, Baim DS, et al. A randomized compar- govern drug distribution for stent-based delivery. Circulation
ison of coronary-stent placement and balloon angioplasty in the 2001; 104:600–605.
treatment of coronary artery disease. Stent Restenosis Study 25. Langer R. New methods of drug delivery. Science 1990; 249:1527–
Investigators. N Engl J Med 1994; 331:496–501. 1533.
7. Praz L, Cook S, Meier B, on behalf of the Working Group on 26. van der Giessen WJ, Lincoff AM, Schwartz RS, et al. Marked
Interventional Cardiology (WG 10) of the European Society of inflammatory sequelae to implantation of biodegradable and
Cardiology/European Association of Percutaneous Cardiovas- nonbiodegradable polymers in porcine coronary arteries. Circu-
cular Interventions (EAPCI). Percutaneous coronary interven- lation 1996; 94:1690–1697.
tions in Europe in 2005. EuroIntervention 2008; 3:442–446. 27. Marx SO, Marks AR. Bench to bedside: the development of
8. Serruys PW, van Hout B, Bonnier H, et al. Randomised compar- rapamycin and its application to stent restenosis. Circulation
ison of implantation of heparin-coated stents with balloon angio- 2001; 104:852–855.
plasty in selected patients with coronary artery disease 28. Poon M, Badimon JJ, Fuster V. Overcoming restenosis with
(Benestent II). Lancet 1998; 352:673–681. sirolimus: from alphabet soup to clinical reality. Lancet 2002;
9. Lev EI, Assali AR, Teplisky I, et al. Comparison of outcomes up 359:619–622.
to six months of heparin-Coated with noncoated stents after 29. Sousa JE, Costa MA, Abizaid AC, et al. Sustained suppression of
percutaneous coronary intervention for acute myocardial infarc- neointimal proliferation by sirolimus-eluting stents: one-year
tion. Am J Cardiol 2004; 93:741–743. angiographic and intravascular ultrasound follow-up. Circula-
10. Kastrati A, Schomig A, Dirschinger J, et al. Increased risk of tion 2001; 104:2007–2011.
restenosis after placement of gold-coated stents: results of a 30. Sousa JE, Costa MA, Sousa AG, et al. Two-year angiographic and
randomized trial comparing gold-coated with uncoated steel intravascular ultrasound follow-up after implantation of siroli-
stents in patients with coronary artery disease. Circulation mus-eluting stents in human coronary arteries. Circulation 2003;
2000; 101:2478–2483. 107:381–383.
11. Rogers C, Edelman ER. Endovascular stent design dictates 31. Morice MC, Serruys PW, Sousa JE, et al. A randomized compar-
experimental restenosis and thrombosis. Circulation 1995; 91: ison of a sirolimus-eluting stent with a standard stent for coro-
2995–3001. nary revascularization. N Engl J Med 2002; 346:1773–1780.
12. Garasic JM, Edelman ER, Squire JC, et al. Stent and artery 32. Holmes DR Jr., Leon MB, Moses JW, et al. Analysis of 1-year
geometry determine intimal thickening independent of arterial clinical outcomes in the SIRIUS trial: a randomized trial of a
injury. Circulation 2000; 101:812–818. sirolimus-eluting stent versus a standard stent in patients at high
13. Lansky AJ, Roubin GS, O’Shaughnessy CD, et al. Randomized risk for coronary restenosis. Circulation 2004; 109:634–640.
comparison of GR-II stent and Palmaz-Schatz stent for elective 33. Kastrati A, Mehilli J, Pache J, et al. Analysis of 14 trials compar-
treatment of coronary stenoses. Circulation 2000; 102:1364–1368. ing sirolimus-eluting stents with bare-metal stents. N Engl J Med
14. Baim DS, Cutlip DE, Midei M, et al. Final results of a randomized 2007; 356:1030–1039.
trial comparing the MULTI-LINK stent with the Palmaz-Schatz 34. Schuler W, Sedrani R, Cottens S, et al. SDZ RAD, a new
stent for narrowings in native coronary arteries. Am J Cardiol rapamycin derivative: pharmacological properties in vitro and
2001; 87:157–162. in vivo. Transplantation 1997; 64:36–42.
15. Baim DS, Cutlip DE, O’Shaughnessy CD, et al. Final results of a 35. Joner M, Nakazawa G, Finn AV, et al. Endothelial cell recovery
randomized trial comparing the NIR stent to the Palmaz-Schatz between comparator polymer-based drug-eluting stents. J Am
stent for narrowings in native coronary arteries. Am J Cardiol Coll Cardiol 2008; 52:333–342.
2001; 87:152–156. 36. Serruys P, Ong A, Piek J, et al. A randomized comparison of a
16. Kastrati A, Hall D, Schomig A. Long-term outcome after coro- durable polymer Everolimus-eluting stent with a bare metal coro-
nary stenting. Curr Control Trials Cardiovasc Med 2000; 1:48–54. nary stent: the SPIRIT first trial. EuroIntervention 2005; 1:58–65.
17. Hill-West JL, Chowdhury SM, Slepian MJ, et al. Inhibition of 37. Serruys P, Ruygrok P, Neuzner J, et al. A randomised compar-
thrombosis and intimal thickening by in situ photopolymeriza- ison of an everolimus-eluting coronary stent with a paclitaxel-
tion of thin hydrogel barriers. Proc Natl Acad Sci U S A 1994; eluting coronary stent: the SPIRIT II trial. EuroIntervention 2006;
91:5967–5971. 2:286–294.
18. West JL, Hubbell JA. Separation of the arterial wall from blood 38. Stone GW, Midei M, Newman W, et al. Comparison of an
contact using hydrogel barriers reduces intimal thickening after everolimus-eluting stent and a paclitaxel-eluting stent in patients
[17/6/010/9:23:2] [412–428]
with coronary artery disease: a randomized trial. JAMA 2008; slow-release paclitaxel-eluting stent for de novo coronary lesions.
299:1903–1913. Circulation 2003; 107:38–42.
39. Meredith I, Ormiston J, Whitbourn R, et al. First-in-human study 57. Colombo A, Drzewiecki J, Banning A, et al. Randomized study to
of the Endeavor ABT-578-eluting phosphorylcholine-encapsu- assess the effectiveness of slow- and moderate-release polymer-
lated stent system in de novonative coronary artery lesions: based paclitaxel-eluting stents for coronary artery lesions. Circu-
Endeavor I Trial. EuroIntervention 2005; 1:157–164. lation 2003; 108:788–794.
40. Fajadet J, Wijns W, Laarman GJ, et al. Randomized, double-blind, 58. Stone GW, Ellis SG, Cox DA, et al. A polymer-based, paclitaxel-
multicenter study of the Endeavor zotarolimus-eluting phos- eluting stent in patients with coronary artery disease. N Engl J
phorylcholine-encapsulated stent for treatment of native coro- Med 2004; 350:221–231.
nary artery lesions: clinical and angiographic results of the 59. Stone GW, Ellis SG, Cannon L, et al. Comparison of a polymer-
ENDEAVOR II trial. Circulation 2006; 114:798–806. based paclitaxel-eluting stent with a bare metal stent in patients
41. Kandzari DE, Leon MB, Popma JJ, et al. Comparison of zotar- with complex coronary artery disease: a randomized controlled
olimus-eluting and sirolimus-eluting stents in patients with trial. JAMA 2005; 294:1215–1223.
native coronary artery disease: a randomized controlled trial. 60. Dawkins KD, Grube E, Guagliumi G, et al. Clinical efficacy of
J Am Coll Cardiol 2006; 48:2440–2447. polymer-based paclitaxel-eluting stents in the treatment of com-
42. Leon MB. Randomized comparison of zotarolimus-eluting and plex, long coronary artery lesions from a multicenter, random-
paclitaxel-eluting stents in patients with coronary artery disease ized trial: support for the use of drug-eluting stents in
(ENDEAVOR IV). In: Transcatheter Cardiovascular Therapeutics contemporary clinical practice. Circulation 2005; 112:3306–3313.
(TCT). Washington, D.C., 2008. 61. Tanabe K, Serruys PW, Grube E, et al. TAXUS III Trial: in-stent
43. Leon MB. Endeavor IV: A randomized comparison of a zotar- restenosis treated with stent-based delivery of paclitaxel incor-
olimus-eluting stent and a paclitaxel-eluting stent in patients porated in a slow-release polymer formulation. Circulation 2003;
with coronary heart disease. In: Transcatheter Cardiovascular 107:559–564.
Therapeutics (TCT). Washington, DC, 2007. 62. Stone GW, Ellis SG, O’Shaughnessy CD, et al. Paclitaxel-eluting
44. Hezi-Yamit A, Sullivan C, Wong J, et al. Impact of polymer stents vs vascular brachytherapy for in-stent restenosis within
hydrophilicity on biocompatibility: implication for DES polymer bare-metal stents: the TAXUS V ISR randomized trial. JAMA
design. J Biomed Mater Res A 2009; 90(1):133–141. 2006; 295:1253–1263.
45. Meredith I, Worthley S, Whitbourn R, et al. The next-generation 63. Lincoff AM, Furst JG, Ellis SG, et al. Sustained local delivery of
EndeavorTM ResoluteTM stent: 4-month clinical and angiographic dexamethasone by a novel intravascular eluting stent to prevent
results from the ResoluteTM first-in-man trial. EuroIntervention restenosis in the porcine coronary injury model. J Am Coll
2007; 3:50–53. Cardiol 1997; 29:808–816.
46. Grube E, Buellesfeld L. BioMatrix Biolimus A9-eluting coronary 64. Tamai H, Igaki K, Kyo E, et al. Initial and 6-month results of
stent: a next-generation drug-eluting stent for coronary artery biodegradable poly-l-lactic acid coronary stents in humans. Cir-
disease. Expert Rev Med Devices 2006; 3:731–741. culation 2000; 102:399–404.
47. Costa RA, Lansky AJ, Abizaid A, et al. Angiographic results of the 65. George BS, Voorhees WD III, Roubin GS, et al. Multicenter
first human experience with the Biolimus A9 drug-eluting stent for investigation of coronary stenting to treat acute or threatened
de novo coronary lesions. Am J Cardiol 2006; 98:443–446. closure after percutaneous transluminal coronary angioplasty:
48. Windecker S, Serruys PW, Wandel S, et al. Biolimus-eluting stent clinical and angiographic outcomes. J Am Coll Cardiol 1993;
with biodegradable polymer versus sirolimus-eluting stent with 22:135–143.
durable polymer for coronary revascularisation (LEADERS): a 66. Lincoff AM, Topol EJ, Chapekis AT, et al. Intracoronary stenting
randomised non-inferiority trial. Lancet 2008; 372:1163–1173. compared with conventional therapy for abrupt vessel closure
49. Chevalier B. NOBORI I Phase II: a prospective, randomized trial complicating coronary angioplasty: a matched case-control
comparing a biolimus A9-eluting stent and a paclitaxel-eluting study. J Am Coll Cardiol 1993; 21:866–875.
stent in patients with coronary artery disease: 9-month clinical 67. Schomig A, Kastrati A, Mudra H, et al. Four-year experience with
and angiographic results. In: Transcatheter Cardiovascular Ther- Palmaz-Schatz stenting in coronary angioplasty complicated by
apeutics (TCT). Washington, D.C., 2007. dissection with threatened or present vessel closure. Circulation
50. Chevalier B, Serruys P, Silber S, et al. Randomised comparison of 1994; 90:2716–2724.
NoboriTM, biolimus A9-eluting coronary stent with a Taxus1, 68. Weaver WD, Reisman MA, Griffin JJ, et al. Optimum percuta-
paclitaxel-eluting coronary stent in patients with stenosis in neous transluminal coronary angioplasty compared with routine
native coronary arteries: the Nobori 1 trial. EuroIntervention stent strategy trial (OPUS-1): a randomised trial. Lancet 2000;
2007; 2:426–434. 355:2199–2203.
51. Hamilos MI, Ostojic M, Beleslin B, et al. Differential effects of 69. Chen MS, John JM, Chew DP, et al. Bare metal stent restenosis is
drug-eluting stents on local endothelium-dependent coronary not a benign clinical entity. Am Heart J 2006; 151:1260–1264.
vasomotion. J Am Coll Cardiol 2008; 51:2123–2129. 70. Doyle B, Rihal CS, O’Sullivan CJ, et al. Outcomes of stent throm-
52. Axel DI, Kunert W, Goggelmann C, et al. Paclitaxel inhibits arterial bosis and restenosis during extended follow-up of patients treated
smooth muscle cell proliferation and migration in vitro and in vivo with bare-metal coronary stents. Circulation 2007; 116:2391–2398.
using local drug delivery. Circulation 1997; 96:636–645. 71. Walters DL, Harding SA, Walsh CR, et al. Acute coronary syn-
53. Heldman AW, Cheng L, Jenkins GM, et al. Paclitaxel stent drome is a common clinical presentation of in-stent restenosis.
coating inhibits neointimal hyperplasia at 4 weeks in a porcine Am J Cardiol 2002; 89:491–494.
model of coronary restenosis. Circulation 2001; 103:2289–2295. 72. Stone GW, Ellis SG, Colombo A, et al. Offsetting impact of
54. Stone GW, Moses JW, Ellis SG, et al. Safety and efficacy of thrombosis and restenosis on the occurrence of death and
sirolimus- and paclitaxel-eluting coronary stents. N Engl J Med myocardial infarction after paclitaxel-eluting and bare metal
2007; 356:998–1008. stent implantation. Circulation 2007; 115:2842–2847.
55. Krucoff MW, Kereiakes DJ, Petersen JL, et al. A novel bioresorb- 73. Weintraub WS, Ghazzal ZM, Douglas JS Jr., et al. Long-term
able polymer paclitaxel-eluting stent for the treatment of single clinical follow-up in patients with angiographic restudy after
and multivessel coronary disease: primary results of the successful angioplasty. Circulation 1993; 87:831–840.
COSTAR (Cobalt Chromium Stent With Antiproliferative for 74. Van Belle E, Ketelers R, Bauters C, et al. Patency of percutaneous
Restenosis) II study. J Am Coll Cardiol 2008; 51:1543–1552. transluminal coronary angioplasty sites at 6-month angiographic
56. Grube E, Silber S, Hauptmann KE, et al. TAXUS I: six- and follow-up: a key determinant of survival in diabetics after coro-
twelve-month results from a randomized, double-blind trial on a nary balloon angioplasty. Circulation 2001; 103:1218–1224.
[17/6/010/9:23:2] [412–428]
75. Schuhlen H, Kastrati A, Mehilli J, et al. Restenosis detected by balloon angioplasty. Restenosis Stent Study Group. N Engl J Med
routine angiographic follow-up and late mortality after coronary 1998; 339:1672–1678.
stent placement. Am Heart J 2004; 147:317–322. 99. Savage MP, Douglas JS Jr., Fischman DL, et al. Stent placement
76. Babapulle MN, Joseph L, Belisle P, et al. A hierarchical Bayesian compared with balloon angioplasty for obstructed coronary
meta-analysis of randomised clinical trials of drug-eluting stents. bypass grafts. Saphenous Vein De Novo Trial Investigators.
Lancet 2004; 364:583–591. N Engl J Med 1997; 337:740–747.
77. Moses JW, Leon MB, Popma JJ, et al. Sirolimus-eluting stents 100. Hanekamp CE, Koolen JJ, Den Heijer P, et al. Randomized study
versus standard stents in patients with stenosis in a native to compare balloon angioplasty and elective stent implantation in
coronary artery. N Engl J Med 2003; 349:1315–1323. venous bypass grafts: the Venestent study. Catheter Cardiovasc
78. Serruys PW, Kutryk MJ, Ong AT. Coronary-artery stents. N Engl Interv 2003; 60:452–457.
J Med 2006; 354:483–495. 101. Sirnes PA, Golf S, Myreng Y, et al. Sustained benefit of stenting
79. Stettler C, Wandel S, Allemann S, et al. Outcomes associated with chronic coronary occlusion: long-term clinical follow-up of the
drug-eluting and bare-metal stents: a collaborative network Stenting in Chronic Coronary Occlusion (SICCO) study. J Am
meta-analysis. Lancet 2007; 370:937–948. Coll Cardiol 1998; 32:305–310.
80. Tu JV, Bowen J, Chiu M, et al. Effectiveness and safety of drug- 102. Buller CE, Dzavik V, Carere RG, et al. Primary stenting versus
eluting stents in Ontario. N Engl J Med 2007; 357:1393–1402. balloon angioplasty in occluded coronary arteries: the Total Occlu-
81. Lagerqvist B, James SK, Stenestrand U, et al. Long-term out- sion Study of Canada (TOSCA). Circulation 1999; 100:236–242.
comes with drug-eluting stents versus bare-metal stents in Swe- 103. Rubartelli P, Verna E, Niccoli L, et al. Coronary stent implanta-
den. N Engl J Med 2007; 356:1009–1019. tion is superior to balloon angioplasty for chronic coronary
82. Jensen LO, Maeng M, Kaltoft A, et al. Stent thrombosis, myo- occlusions: six-year clinical follow-up of the GISSOC trial. J Am
cardial infarction, and death after drug-eluting and bare-metal Coll Cardiol 2003; 41:1488–1492.
stent coronary interventions. J Am Coll Cardiol 2007; 50:463–470. 104. Stone GW, Grines CL, Cox DA, et al. Comparison of angioplasty
83. Finn AV, Joner M, Nakazawa G, et al. Pathological correlates of with stenting, with or without abciximab, in acute myocardial
late drug-eluting stent thrombosis: strut coverage as a marker of infarction. N Engl J Med 2002; 346:957–966.
endothelialization. Circulation 2007; 115:2435–2441. 105. Grines CL, Cox DA, Stone GW, et al. Coronary angioplasty with
84. Joner M, Finn AV, Farb A, et al. Pathology of drug-eluting stents or without stent implantation for acute myocardial infarction.
in humans: delayed healing and late thrombotic risk. J Am Coll Stent Primary Angioplasty in Myocardial Infarction Study
Cardiol 2006; 48:193–202. Group. N Engl J Med 1999; 341:1949–1956.
85. Hofma SH, van der Giessen WJ, van Dalen BM, et al. Indication 106. Antoniucci D, Santoro GM, Bolognese L, et al. A clinical trial
of long-term endothelial dysfunction after sirolimus-eluting stent comparing primary stenting of the infarct-related artery with opti-
implantation. Eur Heart J 2006; 27:166–170. mal primary angioplasty for acute myocardial infarction: results
86. Togni M, Raber L, Cocchia R, et al. Local vascular dysfunction from the Florence Randomized Elective Stenting in Acute Coronary
after coronary paclitaxel-eluting stent implantation. Int J Cardiol Occlusions (FRESCO) trial. J Am Coll Cardiol 1998; 31:1234–1239.
2007; 120:212–220. 107. Spaulding C, Daemen J, Boersma E, et al. A pooled analysis of
87. Togni M, Windecker S, Cocchia R, et al. Sirolimus-eluting stents data comparing sirolimus-eluting stents with bare-metal stents.
associated with paradoxic coronary vasoconstriction. J Am Coll N Engl J Med 2007; 356:989–997.
Cardiol 2005; 46:231–236. 108. Pache J, Dibra A, Mehilli J, et al. Drug-eluting stents compared
88. Cook S, Wenaweser P, Togni M, et al. Incomplete stent apposi- with thin-strut bare stents for the reduction of restenosis: a
tion and very late stent thrombosis after drug-eluting stent prospective, randomized trial. Eur Heart J 2005; 26:1262–1268.
implantation. Circulation 2007; 115:2426–2434. 109. Brunner-La Rocca HP, Kaiser C, Pfisterer M. Targeted stent use
89. Feres F, Costa JR Jr., Abizaid A. Very late thrombosis after drug- in clinical practice based on evidence from the Basel Stent Cost
eluting stents. Catheter Cardiovasc Interv 2006; 68:83–88. Effectiveness Trial (BASKET). Eur Heart J 2007; 28:719–725.
90. Mintz GS, Weissman NJ. Intravascular ultrasound in the drug- 110. Hoye A, Iakovou I, Ge L, et al. Long-term outcomes after stenting
eluting stent era. J Am Coll Cardiol 2006; 48:421–429. of bifurcation lesions with the “crush” technique: predictors of an
91. Stahli BE, Camici GG, Steffel J, et al. Paclitaxel enhances throm- adverse outcome. J Am Coll Cardiol 2006; 47:1949–1958.
bin-induced endothelial tissue factor expression via c-Jun termi- 111. Steigen TK, Maeng M, Wiseth R, et al. Randomized study on simple
nal NH2 kinase activation. Circ Res 2006; 99:149–155. versus complex stenting of coronary artery bifurcation lesions: the
92. Steffel J, Latini RA, Akhmedov A, et al. Rapamycin, but not FK- Nordic bifurcation study. Circulation 2006; 114:1955–1961.
506, increases endothelial tissue factor expression: implications 112. Steinhubl SR, Berger PB, Mann JT III, et al. Early and sustained
for drug-eluting stent design. Circulation 2005; 112:2002–2011. dual oral antiplatelet therapy following percutaneous coronary
93. Moreno R, Fernandez C, Hernandez R, et al. Drug-eluting stent intervention: a randomized controlled trial. JAMA 2002;
thrombosis: results from a pooled analysis including 10 random- 288:2411–2420.
ized studies. J Am Coll Cardiol 2005; 45:954–959. 113. Pfisterer M, Brunner-La Rocca HP, Buser PT, et al. Late clinical
94. Mauri L, Hsieh WH, Massaro JM, et al. Stent thrombosis in events after clopidogrel discontinuation may limit the benefit of
randomized clinical trials of drug-eluting stents. N Engl J Med drug-eluting stents: an observational study of drug-eluting ver-
2007; 356:1020–1029. sus bare-metal stents. J Am Coll Cardiol 2006; 48:2584–2591.
95. Buring JE, Glynn RJ, Hennekens CH. Calcium channel blockers 114. Eisenstein EL, Anstrom KJ, Kong DF, et al. Clopidogrel use and
and myocardial infarction. A hypothesis formulated but not yet long-term clinical outcomes after drug-eluting stent implanta-
tested. JAMA 1995; 274:654–655. tion. JAMA 2007; 297:159–168.
96. Versaci F, Gaspardone A, Tomai F, et al. A comparison of 115. Airoldi F, Colombo A, Morici N, et al. Incidence and predictors
coronary-artery stenting with angioplasty for isolated stenosis of drug-eluting stent thrombosis during and after discontinua-
of the proximal left anterior descending coronary artery. N Engl J tion of thienopyridine treatment. Circulation 2007; 116:745–754.
Med 1997; 336:817–822. 116. Daemen J, Wenaweser P, Tsuchida K, et al. Early and late
97. Betriu A, Masotti M, Serra A, et al. Randomized comparison of coronary stent thrombosis of sirolimus-eluting and paclitaxel-
coronary stent implantation and balloon angioplasty in the treat- eluting stents in routine clinical practice: data from a large two-
ment of de novo coronary artery lesions (START): a four-year institutional cohort study. Lancet 2007; 369:667–678.
follow-up. J Am Coll Cardiol 1999; 34:1498–1506. 117. Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versus
98. Erbel R, Haude M, Hopp HW, et al. Coronary-artery stenting clopidogrel in patients with acute coronary syndromes. N Engl J
compared with balloon angioplasty for restenosis after initial Med 2007; 357:2001–2015.
[17/6/010/9:23:2] [412–428]
118. Grines CL, Bonow RO, Casey DE Jr., et al. Prevention of prema- elevation myocardial infarction: the importance of thrombus
ture discontinuation of dual antiplatelet therapy in patients with burden. J Am Coll Cardiol 2007; 50:573–583.
coronary artery stents: a science advisory from the American 122. Kastrati A, Dibra A, Spaulding C, et al. Meta-analysis of random-
Heart Association, American College of Cardiology, Society for ized trials on drug-eluting stents vs. bare-metal stents in patients
Cardiovascular Angiography and Interventions, American Col- with acute myocardial infarction. Eur Heart J 2007; 28:2706–2713.
lege of Surgeons, and American Dental Association, with repre- 123. Wilson SH, Fasseas P, Orford JL, et al. Clinical outcome of
sentation from the American College of Physicians. Circulation patients undergoing non-cardiac surgery in the two months
2007; 115:813–818. following coronary stenting. J Am Coll Cardiol 2003; 42:234–240.
119. Hodgson JM, Stone GW, Lincoff AM, et al. Late stent thrombosis: 124. Vicenzi MN, Meislitzer T, Heitzinger B, et al. Coronary artery
considerations and practical advice for the use of drug-eluting stenting and non-cardiac surgery—a prospective outcome study.
stents: a report from the Society for Cardiovascular Angiography Br J Anaesth 2006; 96:686–693.
and Interventions Drug-eluting Stent Task Force. Catheter Car- 125. Smith SC Jr., Feldman TE, Hirshfeld JW Jr., et al. ACC/AHA/
diovasc Interv 2007; 69:327–333. SCAI 2005 guideline update for percutaneous coronary inter-
120. Urban P, Gershlick AH, Guagliumi G, et al. Safety of coronary vention: a report of the American College of Cardiology/Amer-
sirolimus-eluting stents in daily clinical practice: one-year fol- ican Heart Association Task Force on Practice Guidelines (ACC/
low-up of the e-Cypher registry. Circulation 2006; 113:1434–1441. AHA/SCAI Writing Committee to Update 2001 Guidelines for
121. Sianos G, Papafaklis MI, Daemen J, et al. Angiographic stent Percutaneous Coronary Intervention). Circulation 2006; 113:
thrombosis after routine use of drug-eluting stents in ST-segment e166–e286.
[17/6/010/10:7:28] [429–443]
33
Primary PCI in ST elevation myocardial infarction

Sean P. Javaheri and James E. Tcheng
INTRODUCTION reperfusion” (25). These issues contributed to the search for

Perspectives alternative rapid, complete, and sustained approaches to reper-
Every year, millions of individuals throughout the world sus- fusion in STEMI.
tain an acute myocardial infarction (AMI). The vast majority of
these events are initiated by disruption or erosion of the thin
fibrous cap of a vulnerable atherosclerotic plaque, which results
Angioplasty Awakenings
Contemporaneous with the initial explorations of fibrinolytic
in the exposure of subendothelial constituents of the arterial
therapy in STEMI, Andreas Gruentzig was introducing the
wall and the consequent triggering of the thrombosis cascade
world to coronary balloon angioplasty (26,27). In the early
(1–4). Appreciation of this pathophysiological construct, accom-
days, the potential applications were not just uncertain—they
panied by the recognition that coronary artery occlusion could
were undiscovered. The ability to mechanically reperfuse an
generally be identified by ST segment elevation on the electro-
acutely occluded coronary artery was demonstrated by Peter
cardiogram (5), were the key insights prompting the explora-
Rentrop in 1979; he reestablished flow in a right coronary artery
tion of reperfusion strategies. Further evolution of our
that abruptly occluded during diagnostic angiography by pass-
understanding of the spectrum of disease now encompassed
ing a coronary guidewire through the occlusion (28). The first
by the term acute coronary syndrome (ACS) (6) resulted in
primary PCI is generally attributed to Geoffrey Hartzler for a
abandoning the terms “non–Q wave” and “Q-wave” myocar-
balloon angioplasty procedure performed in 1980 on a patient
dial infarction in favor of the current lexicon of “unstable
sustaining an acute inferior STEMI due to occlusion of a high-
angina/non–ST elevation myocardial infarction (UA/
grade right coronary lesion that had been documented by
NSTEMI)” and “ST elevation myocardial infarction (STEMI).”
diagnostic angiography the day before (29). The first published
The focus of this chapter is primary percutaneous coronary
account appeared shortly thereafter, not as a scientific report,
intervention (PCI), including the processes of emergency refer-
but as a first person account in a local newspaper, of a primary
ral for diagnostic cardiac catheterization and infarct-related
PCI performed by Robert “Jess” Peter at Duke University
artery PCI as the preferred approach to managing the patient
Medical Center on a columnist for the Durham Morning Herald
with STEMI (7).
(30). Anecdotally, others began performing primary PCI on the
The description by Rentrop and colleagues that the
occasional patient brought to the catheterization suite present-
fibrinolytic agent streptokinase could restore coronary perfu-
ing with STEMI, not necessarily as an intentioned treatment but
sion in STEMI was the seminal event that ushered in the
more as one of opportunity.
reperfusion therapy era (8). The benefits of fibrinolytic therapy
are well documented in the randomized clinical trials literature
(9–18). Overall, it has been shown that 30 deaths are prevented Fibrinolysis Vs. Primary PCI
for every 1000 patients treated with fibrinolysis (19). The great- These early successes led to the design and conduct of three
est benefit of fibrinolytic therapy is seen in the first six hours of landmark trials of primary PCI versus fibrinolysis published as
symptom onset and is comparable to primary PCI when given adjoining articles in the March 11, 1993 edition of the New
within two to three hours of symptom onset. Efficacy increases England Journal of Medicine (31–33). The principal findings of
with patient risk, with patients with new (or presumed new) these pioneering studies—reliable acute restoration of myocar-
left bundle branch block deriving the greatest benefit, followed dial perfusion in over 80% of patients, with the near elimination
by those with anterior wall myocardial infarction; conversely, of intracerebral hemorrhage, resulting in improved clinical
the least benefit is seen in those with an uncomplicated inferior outcomes—have proven remarkably durable. Since then, over
wall MI, with only eight lives saved per 1000 patients treated. 20 comparative randomized trials of primary PCI versus fibri-
However, treatment with fibrinolytic therapy is not with- nolysis have been published. A 2003 meta-analysis of these
out limitations. A prominent hazard is the increased risk of trials provides a fairly reliable estimate of the improvement in
hemorrhagic stroke, with most of the hazard appearing on the outcomes seen with primary PCI compared with fibrinolysis,
day of treatment (9). The rate of intracranial hemorrhage including an approximate 2% absolute (25% relative) reduction
approaches 1% and is associated with a 60% fatality rate (20–23). in short-term mortality (7.4% vs. 5.3%, p ¼ 0.0003; Fig. 33.1)
Other types of major, life-threatening bleeding occur in up to (34). Interestingly, these salutary effects approximate the rela-
4% to 13% of patients (21,22,24). Were these the only issues, tive mortality reduction of fibrinolysis compared with placebo;
fibrinolysis might still be preeminent; however, complete reper- in the nine large randomized clinical trials of fibrinolysis versus
fusion, the actual goal of therapy, is achieved in perhaps only placebo, mortality at 35 days was also reduced by approxi-
25% of patients treated with fibrinolysis, termed the “illusion of mately 2% (11.5% vs. 9.6%, p < 0.000,01) (Fig. 33.2) (9,34).
[17/6/010/10:7:28] [429–443]
(50% vs. 63%, p ¼ 0.03) even though the enrollment goal was
not met. The survival advantage was most pronounced in
patients treated within 6 hours of symptom onset, although
the median time to treatment was approximately 12 hours. At
six years, there was a 13.2% absolute increase in survival with
early revascularization compared with initial medical stabili-
zation (32.8 vs. 19.6%, p ¼ 0.03) (39). Among hospital survivors
(n ¼ 143), there was also a sustained mortality benefit with
early revascularization compared with immediate medical sta-
bilization (62.4 vs. 44.4%, p ¼ 0.03).
Despite the overall salutary benefit, there was a trend
toward a mortality hazard with early revascularization in those
older than 75 years of age (75.0% with primary PCI vs. 53.1%
with medical management, p ¼ NS). This contributed to the
2004 ACC/AHA guidelines class IIa recommendation for pri-
Figure 33.1 Results of a meta-analysis of 23 randomized trials mary PCI in the elderly versus a class I indication for those
comparing short-term outcomes of primary PCI versus fibrinolytic younger than 75 years (19). Subsequently, a possible explana-
therapy in acute ST elevation myocardial infarction. Abbreviation: tion for the divergent results regarding age has been published
PCI, percutaneous coronary intervention. Source: Adapted from (40). As might be anticipated, in the SHOCK trial, the 30-day
Ref. 34. mortality rate in the early revascularization group was higher
in the elderly (75.0%) than in those less than 75 (41.4%). How-
ever, there were significant imbalances between the emergency
revascularization elderly cohort and the initial medical stabili-
zation elderly cohort; specifically, there were greater propor-
tions of women and those with an anterior myocardial
infarction, and the baseline ejection fraction was lower in the
emergency revascularization group. Furthermore, the 30-day
mortality was 53.1% in the elderly managed medically versus
56.8% in younger patients. Also, the number of patients in the
elderly subgroup (56 patients) was small. The authors argued
that since there was a numerically lower mortality in patients
older than 75 years in the immediate medical stabilization arm,
differences in baseline characteristics (and the small sample
size) were more likely to be responsible for the possible hazard
of early revascularization in the elderly than an attributable
effect of treatment.
Results of the Senior Primary Angioplasty in Acute
Figure 33.2 In-hospitality mortality in 9 trials that compared fibri- Myocardial Infarction (SENIOR PAMI) trial directly addressed
nolytic therapy with controls (left) and in 23 subsequent trials of the issue of primary PCI in the elderly (41). In the SENIOR
primary PCI versus fibrinolytic therapy (right) in ST elevation acute PAMI trial, 481 patients greater than 70 years old were random-
myocardial infarction. Abbreviation: PCI, percutaneous coronary
ized to primary PCI or fibrinolytic therapy, with the primary
intervention. Adapted from Refs. 9 and 34.
end point being death or disabling stroke. At 30 days, while
there was no significant difference in mortality (10% vs. 13%,
p ¼ 0.48) or in the combined end point of death or disabling
stroke (11.3% vs. 13%, p ¼ 0.57), there was a reduction in the
INDICATIONS composite end point of death, cerebrovascular accident, and
Updated Lessons from Clinical Trials reinfarction (11.6% vs. 18%, p ¼ 0.05). Perhaps more impor-
Generically, primary PCI has proven as good as or better than tantly, there was no suggestion of harm in the elderly with
fibrinolysis (35,36). In an updated meta-analysis of the random- primary PCI.
ized trials comparing fibrinolysis with primary PCI conducted
from 1990 to 2006, there remained highly statistically significant
relative risk reductions in mortality (28%), recurrent MI (63%), Transfer for Primary PCI
and stroke (50%) (37). With this continuing experience, suffi- The first decade of primary PCI versus fibrinolysis investiga-
cient numbers accrued to address subgroups where there tion was conducted primarily at tertiary centers capable of
remained the question of harm with primary PCI: patients performing on-site primary PCI. Following the demonstration
presenting in cardiogenic shock and the elderly. of efficacy and safety of primary PCI in this setting, the concept
In the Should We Emergently Revascularize Occluded of transferring patients with STEMI for primary PCI became the
Coronaries for Cardiogenic Shock (SHOCK) trial, 302 patients next logical step. Two seminal trials heralded the now common
presenting in cardiogenic shock were randomized to emergency approach of transferring patients acutely for primary PCI when
revascularization or initial medical stabilization (38). The unset- PCI can be accomplished promptly. The Danish Trial in Acute
tling finding of the SHOCK trial was the high 30-day mortality Myocardial Infarction 2 (DANAMI 2) (42) and the Primary
seen in both cohorts. However, the six-month results revealed Angioplasty in Patients Transferred from General Community
a significant reduction in mortality with revascularization Hospitals to Specialized PTCA Units with or Without
[17/6/010/10:7:28] [429–443]
PRIMARY PCI IN ST ELEVATION MYOCARDIAL INFARCTION 431
Emergency Thrombolysis 2 (PRAGUE 2) (43) trials compared

the outcomes of STEMI patients undergoing primary PCI at a
tertiary hospital (including those transferred from an initial
hospital without primary PCI capabilities) with those receiving
fibrinolysis and managed more electively. In the DANAMI 2
trial, of the 1572 patients enrolled, 567 were randomized to
transfer for primary PCI while 562 received fibrinolysis at the
initial referring hospital; another 220 patients were randomized
to fibrinolysis at the tertiary center and 223 were referred for
emergency primary PCI. The primary end point was the com-
posite of death, reinfarction, or stroke at 30 days, and occurred
less frequently in the primary PCI group than in the fibrinolysis
group (8.5% vs. 14.2%, p ¼ 0.002). Although trends favoring
primary PCI were seen in mortality and stroke, the observed
efficacy was primarily due to a reduction in reinfarction (1.6%
vs. 6.3%, p < 0.001). It should be noted that in DANAMI 2,
transfer times were quite short, contributing to rapid door-to- Figure 33.3 Results of a pooled analysis of eight prospective
randomized trials of patients presenting at noninvasive centers,
balloon (DB) times in the transfer for primary PCI group, and
evaluating short-term outcomes of transfer to a tertiary center for
rescue or elective PCI were discouraged. In this group, the
primary PCI versus on-site fibrinolytic therapy in acute ST elevation
median time from initial presentation to treatment start time myocardial infarction. Abbreviation: PCI, percutaneous coronary
was 117 minutes. In the referring hospital fibrinolysis arm, the intervention. Source: Adapted from Ref. 46.
median time from admission to initiation of fibrinolysis was
64 minutes. The difference between the median time for pri-
mary PCI treatment and time to fibrinolysis administration was
thus only 53 minutes, highlighting the importance of well- longer transfer times and/or DB times remains to be demon-
organized and rapid systems to facilitate transfer of STEMI strated. Of note, in the National Registry of Myocardial Infarc-
patients for primary PCI. tion (NRMI) 3/4 registry in the United States, patients
In the PRAGUE 2 trial, 850 patients were randomized to undergoing transfer for STEMI average DB times of 53 minutes
either immediate fibrinolysis at a referring hospital (n ¼ 421) or at the receiving hospital and have a median interval from
transfer for primary PCI (n ¼ 429) (43). In the transfer for PCI initial hospital arrival to balloon inflation of approximately
group, the median time from randomization to balloon activa- 180 minutes (44).
tion was 97 27 minutes, and in the fibrinolysis arm, the median In 2003, Dalby and colleagues performed a meta-analysis
time from randomization to needle time was 12 10 minutes. of the six published randomized trials of 3750 patients evaluat-
The primary end point in this trial was 30-day mortality, with ing transfer for primary PCI versus immediate fibrinolysis (45).
the trend favoring the primary PCI group (6.8% vs. 10.0%, The primary end point was the composite of death, reinfarction
p ¼ 0.12). The authors performed a nonprespecified mortality and stroke. The time from randomization to primary PCI was
analysis of the 380 patients who actually underwent primary PCI between 80 to 122 minutes. The authors found a 42% relative
versus the 419 patients who received fibrinolysis (five of whom risk reduction in the primary end point (p < 0.001) in the group
were assigned to the PCI group) and found a statistically signif- transferred for primary PCI compared with the group receiving
icant reduction in 30-day mortality (6.0% vs. 10.4%, p < 0.05). on-site fibrinolysis. When the individual components of the
Among the 299 patients randomized greater than three hours composite end points were evaluated, reinfarction was signifi-
after the onset of symptoms, the mortality even more strongly cantly reduced by 68% (p < 0.001) and stroke by 56% (p ¼
favored primary PCI (6.0% with primary PCI vs. 15.3% with 0.015). There was a trend toward reduction in all-cause mor-
fibrinolysis, p < 0.02). Conversely, in patients presenting within tality of 19% (p ¼ 0.086) with primary PCI. The authors
three hours of symptom onset (n ¼ 551), there was no difference concluded that even when transfer for primary PCI is required,
in mortality in the two treatment arms (7.3% with primary PCI primary PCI remained superior to a strategy of immediate
vs. 7.4% with fibrinolysis). fibrinolysis, provided that ambulance systems, prehospital
Important considerations when evaluating the DANAMI 2 management, and appropriate primary PCI capabilities are in
and PRAGUE 2 trials include the use of tissue plasminogen place to rapidly and effectively accomplish the transfer and
activator (tPA) as the fibrinolytic agent in DANAMI 2 and the perform emergency PCI. An updated analysis of an expanded
use of streptokinase in PRAGUE 2. In addition, the rate of stent data set (n ¼ 4155) has provided similar results (Fig. 33.3) (46).
implantation in DANAMI 2 was 93%, whereas in PRAGUE 2 it
was only 63%. Both trials had very low rates of complications
during transfer, with two deaths and three cases of successfully Door-to-Balloon Time Vs. Door-to-Needle Time
resuscitated ventricular fibrillation in PRAGUE 2 and no deaths Given the extensive body of evidence, primary PCI has become
and eight cases of successfully resuscitated ventricular fibrilla- the standard of care for the patient presenting with STEMI.
tion in DANAMI 2. Both trials, while demonstrating the feasi- However, constrained primarily by accessibility, primary PCI is
bility and efficacy of transferring patients for primary PCI, were not universally available, contributing to a continuing debate
performed in small countries with experienced catheterization regarding transfer for primary PCI versus immediate fibrinolysis.
laboratories. The time for transfer was on average 32 minutes in For primary PCI, the current ACC/AHA STEMI guidelines
DANAMI 2 and 48 minutes in PRAGUE 2, with DB times at the identify a target time of less than 90 minutes from presentation
receiving hospitals averaging 26 minutes in both studies. to first treatment device activation; for fibrinolysis, the target for
Whether these results can be extended to environments with time from presentation to administration of thrombolytic therapy
[17/6/010/10:7:28] [429–443]
is less than 30 minutes (19,35,36). At an institutional level, when

the difference between DB and door-to-needle (DN) time can be
expected to exceed 60 minutes, immediate fibrinolysis is there-
fore suggested as the preferred strategy.
In an analysis conducted in 2003 by Nallamothu and
Bates of data compiled from the 23 trials included in the
quantitative review by Keeley, Boura and Grines (34), the
authors evaluated the four- to six-week incidence of death,
reinfarction, and stroke, assessing the effects of PCI-related
time delays on outcomes, where PCI-related time delay was
defined as the difference between DB and DN times (47). The
analysis, conducted utilizing a variance-weighted linear regres-
sion model, evaluated mortality in 21 of the studies, and the
combined end point of death, reinfarction, or stroke in the
13 studies reporting these respective end points. As the PCI-
related time delay increased, the mortality reduction favoring
primary PCI decreased significantly (0.94% decrease for every
additional 10-minute delay, p ¼ 0.006). Overall, the two reper- Figure 33.4 Multivariable analysis of mortality outcomes of
fusion strategies appeared to reach equipoise (where fibrinol- 192,509 patients at 645 National Registry of Myocardial Infarction
ysis outcomes become equal to primary PCI) with regard to hospitals estimating the overall effect of reperfusion with primary
PCI versus fibrinolysis on the basis of increasing PCI-related delay.
mortality once the PCI-related time delay reached 62 minutes
Abbreviations: PCI, percutaneous coronary intervention; DB, door-
or more. Similarly, the impact of the PCI-related time delay on to-balloon; DN, door-to-needle. Source: From Ref. 49.
the composite end point of death, reinfarction, or stroke was
also significant (1.17% decrease for every 10-minute delay,
p ¼ 0.016, time to equipoise of 93 minutes).
There has subsequently been refinement of the analyses
of Nallamothu and Bates. A 2005 reanalysis of the same data by tation of patients directly to facilities capable of primary PCI
Betriu and Masotti utilized a different statistical model to assess (bypassing facilities that do not provide 24/7 coverage) (50–53).
the relationship between PCI-related delay and treatment ben- Where primary PCI cannot be delivered, the focus should be on
efit (48). Magnitude and statistical significance of the relation- reducing delays to fibrinolysis (53). The balance of these sys-
ship were estimated using weighted least-squares regression, in tems-based approaches to reducing DB and DN time will thus
which results from each trial were weighted by the square root influence the optimal strategy of a given institution. Regardless,
of the number of patients in each trial. Linear regression the message is that delays should be reduced to the absolute
analysis was used to estimate the decrease in benefit for each minimum given the uniform improvements in outcomes asso-
additional minute of PCI-related time delay and the delay ciated with decreasing time to treatment.
duration predicted to nullify the benefit of a primary PCI
approach. The key finding remained that the survival benefit
PHARMACOLOGY
of primary PCI over fibrinolysis decreased as the PCI-related
Antithrombotic Therapies
time delay increased, specifically 0.24% for every additional
As STEMI is primarily a thrombosis-mediated event, (non-
10-minute delay. The overall point of equipoise was calculated
fibrinolytic) antithrombotic agents have logically been the sub-
to be 110 minutes, with shorter times to equipoise associated
ject of extensive evaluations as adjuncts to reperfusion
with fibrin-specific fibrinolytic agents, shorter duration of
strategies (7). Studies have included agents directed at mitigat-
symptoms, younger age, and other variables.
ing platelet activity (aspirin, the thienopyrdines clopidogrel
These findings have been confirmed in an expanded anal-
and prasugrel, and the glycoprotein IIb/IIIa inhibitors abcix-
ysis of the NRMI database by Pinto and colleagues (49). In this
imab, eptifibatide, and tirofiban), and inhibiting the coagula-
analysis, relationships between PCI-related delay and in-hospital
tion cascade (heparin, enoxaparin, bivalirudin, and
mortality of 192,509 STEMI patients were evaluated using a
fondaparinux). What has been consistently demonstrated is
hierarchical model to simultaneously adjust for patient-level
that antithrombotic therapies effectively and efficaciously
risk factors and hospital covariates. These investigators identi-
improve outcomes in the acute STEMI setting; what remains
fied a DB minus DN equipoise time of 114 minutes (Fig. 33.4), a
less well-defined is whether there is an optimal approach or a
value remarkably similar to the finding of 110 minutes by Betriu
series of approaches that best balance efficacy and safety in
and Masotti described above. In addition, this analysis also
primary PCI. Reflecting the central import of both pharmaco-
recognized that the point of equipoise varied depending
therapy and primary PCI in the management of STEMI, in 2009
on baseline characteristics. Specifically, it increased with age
the ACC/AHA Task Force on Practice Guidelines authored a
(71 minutes for age <65 vs. 155 minutes for age 65) and duration
joint update combining recommendations for STEMI manage-
of symptoms (94 minutes for symptom duration of 120 minutes
ment and primary PCI into one document (36). A summary of
vs. 190 minutes for symptom duration of >120 minutes). In
the resulting joint guidelines regarding antithrombotic thera-
summary, these analyses have informed the guidelines regard-
pies relevant to primary PCI is provided in Table 33.1.
ing the optimal approach an institution should follow with
STEMI patients. Where primary PCI is available, systems
should focus on strategies to reduce delays such as prehospital Thienopyridines
assessment with 12-lead electrocardiography, activation of There are no pivotal trials explicitly examining the efficacy and
interventional cardiology services from the field, and transpor- safety of clopidogrel (and in particular, a 600-mg loading dose
[17/6/010/10:7:28] [429–443]
Table 33.1 Compilation of 2009 ACC/AHA Recommendations for the Use of Antiplatelet and Anticoagulant Therapies in Primary
Percutaneous Coronary Intervention
Recommendations for the use of glycoprotein IIb/IIIa receptor antagonists
Class IIa
It is reasonable to start treatment with glycoprotein IIb/IIIa receptor antagonists (abciximab level of evidence A; tirofiban level of evidence B;
or eptifibatide level of evidence B) at the time of primary PCI (with or without stenting) in selected patients with STEMI.
Class IIb
The usefulness of glycoprotein IIb/IIIa receptor antagonists (as part of a preparatory pharmacological strategy for patients with STEMI
before their arrival in the cardiac catheterization laboratory for angiography and PCI) is uncertain (level of evidence B).
Recommendations for the use of thienopyridines
Class I
1. A loading dose of thienopyridine is recommended for STEMI patients for whom PCI is planned. Regimens should be one of the following:
a. At least 300–600 mg of clopidogrel should be given as early as possible before or at the time of primary or nonprimary PCI (level of
evidence C).
b. Prasugrel 60 mg should be given as soon as possible for primary PCI (level of evidence B).
2. The duration of thienopyridine therapy should be as follows:
a. In patients receiving a stent (bare-metal or drug-eluting) clopidogrel 75 mg daily (level of evidence B) or prasugrel 10 mg daily (level
of evidence B) should be given for at least 12 mo.
b. If the risk of morbidity because of bleeding outweighs the anticipated benefit afforded by thienopyridine therapy, earlier discontinuation
should be considered (level of evidence C).
Class III
In STEMI patients with a prior history of stroke or transient ischemic attack for whom primary PCI is planned, prasugrel is not recommended
as part of a dual antiplatelet therapy regimen (level of evidence C).
Recommendations for the use of parenteral anticoagulants
Class I
1. For patients proceeding to primary PCI who have been treated with ASA and a thienopyridine, recommended supportive anticoagulant
regimens include the following:
a. For prior treatment with unfractionated heparin, additional boluses of unfractionated heparin should be administered as needed
to maintain therapeutic activated clotting time levels, taking into account whether glycoprotein IIb/IIIa receptor antagonists have been
administered (level of evidence C).
b. Bivalirudin is useful as a supportive measure for primary PCI with or without prior treatment with unfractionated heparin (level
of evidence B).
c. For prior treatment with enoxaparin, if the last subcutaneous dose was administered at least 8–12 hr earlier, an intravenous dose
of 0.3 mg/kg of enoxaparin should be given; if the last subcutaneous dose was administered within the prior 8 hr, no additional
enoxaparin should be given (level of evidence B).
d. For prior treatment with fondaparinux, administer additional intravenous treatment with an anticoagulant possessing anti-IIa
activity, taking into account whether GP IIb/IIIa receptor antagonists have been administered (level of evidence C).
Class IIa
In STEMI patients undergoing PCI who are at high risk of bleeding, bivalirudin anticoagulation is reasonable (level of evidence B).
Abbreviations: PCI, percutaneous coronary intervention; STEMI, ST elevation myocardial infarction.
of clopidogrel) as a direct adjunct to primary PCI. With respect was 3.2 days in the clopidogrel arm and 2.9 days in the placebo
to PCI in STEMI, in the Clopidogrel as Adjunctive Reperfusion arm; these results are thus of only modest relevance to the
Therapy (CLARITY TIMI-28) trial, clopidogrel was compared primary PCI context. Nonetheless, the current STEMI guidelines
with placebo in 3491 patients treated with aspirin, heparin, and for primary PCI recommend a 300- to 600-mg dose of clopidogrel
fibrinolysis (54). The primary end point of the trial, the com- before or at the time of primary PCI (Table 33.2) (36).
posite of death, recurrent myocardial infarction, or an occluded Of perhaps greater relevance are the results observed in
infarct-related artery, before delayed elective angiography, was the STEMI subgroup of patients of the Trial to Assess Improve-
improved in the clopidogrel arm (15% vs. 21.7%, p < 0.001); ment in Therapeutic Outcomes by Optimizing Platelet Inhibi-
the end point was driven primarily by an improvement in tion with Prasugrel (TRITON TIMI-38) trial (58). In the TRITON
the rate of infarct artery patency. With respect to PCI, of the TIMI-38 study, 13,608 patients were randomized to prasugrel or
3491 patients in CLARITY TIMI-28, 1863 underwent coronary clopidogrel therapy as an adjunct to PCI. Of these, 3534 (26%)
intervention (55). In the PCI cohort, there was a significant were in the STEMI cohort, with a large majority of these
reduction in the composite end point of cardiovascular death, managed with primary PCI (59). Prasugrel was administered
myocardial infarction, and stroke with clopidogrel (3.6 vs. 6.2%, as a 60-mg loading dose followed by a 10-mg/day maintenance
p ¼ 0.008), and an independent reduction in cardiovascular dose, while clopidogrel was given as a 300-mg loading dose
death and myocardial infarction (3.3 vs. 5.4%, p ¼ 0.03). How- followed by 75 mg/day. Treatment could be initiated before,
ever, the median number of days from randomization to PCI during, or immediately following PCI. In the STEMI subgroup,
[17/6/010/10:7:28] [429–443]
Table 33.2 Grading Scales for Epicardial Coronary Blood Flow and Microvascular Perfusion per the TIMI Group: the TIMI Coronary Flow
Scale and the TIMI Myocardial Perfusion Scale
TIMI coronary flow scale
TIMI grade 0: No antegrade flow beyond the point of occlusion.
TIMI grade 1: Contrast material passes beyond the area of obstruction but fails to opacify the entire coronary bed distal to the obstruction
for the duration of the cineangiographic filming sequence.
TIMI grade 2: Contrast material passes across the obstruction and opacifies the coronary artery distal to the obstruction. However, the
rate of entry of contrast material into the vessel distal to the obstruction or its rate of clearance from the distal bed (or both) is perceptibly
slower than its flow into or clearance from comparable areas not perfused by the previously occluded vessel.
TIMI grade 3: Antegrade flow into the bed distal to the obstruction occurs as promptly as antegrade flow into the bed proximal to the
obstruction, and clearance of contrast material from the involved bed is as rapid as clearance from an uninvolved bed in the same vessel
or the opposite artery.
TMP scale
TMP grade 0: No apparent tissue-level perfusion (no ground glass appearance of blush or opacification of the myocardium).
TMP grade 1: Presence of a myocardial blush, but without clearance from the microvasculature (blush or stain present on the next injection).
TMP grade 2: Blush clears slowly (blush is strongly persistent and diminishes minimally or not at all during 3 cardiac cycles of the washout
phase).
TMP grade 3: Blush begins to clear during washout (blush is minimally persistent after 3 cardiac cycles of washout).
Abbreviations: TIMI, thrombolysis in myocardial infarction; TMP, TIMI myocardial perfusion.
a 21% relative risk reduction (2.4% absolute risk reduction) in the 4.1% in controls, p ¼ 0.36). It should be acknowledged, how-
primary composite end point of cardiovascular death, nonfatal ever, that most of these trials were conducted before dual
myocardial infarction, or nonfatal stroke at 14.5 months favoring antiplatelet therapy (aspirin plus a thienopyridine) had become
prasugrel was observed (Kaplan-Meier estimate of 12.4% vs. a standard of care in stent PCI.
10.0%, p ¼ 0.02). The efficacy improvement was largely related to More recently, several trials have evaluated platelet gly-
a reduction in nonfatal myocardial infarction in the prasugrel coprotein IIb/IIIa blockade with concomitant dual antiplatelet
arm (6.7% vs. 8.8%, p < 0.02) with a trend favoring mortality (aspirin plus clopidogrel) therapy in the primary PCI setting. In
(2.4% vs. 3.3%, p ¼ NS). Of note, efficacy was greater than had the second Ongoing Tirofiban in Myocardial Infarction Evalu-
been projected; the trial had not been independently powered to ation (ONTIME) 2 trial, 984 acute STEMI patients were
demonstrate efficacy in the STEMI cohort (60). Furthermore, randomized to higher dose tirofiban (25-mcg/kg bolus with a
prasugrel has a more rapid onset of action than clopidogrel 0.15-mcg/kg-min infusion) or placebo at first medical contact in
(61) and thus may be better suited to the emergency primary PCI the ambulance or other prehospital setting (64). All patients
context, particularly when given prior to arrival in the cardiac were given aspirin and a 600-mg loading dose of clopidogrel.
catheterization suite. On the basis of TRITON TIMI-38, prasugrel The primary end point, mean residual ST segment deviation
treatment is now listed as a IB recommendation and an alterna- 60 minutes following primary PCI, was improved in the tirofiban
tive to clopidogrel as an adjunct to primary PCI in STEMI, unless arm (3.6-mm residual deviation with tirofiban vs. 4.8 mm with
contraindicated (Table 33.2) (36). placebo, p ¼ 0.003). Clinically, the key secondary end point of the
composite of death, recurrent myocardial infarction, urgent target
vessel revascularization, or blinded bailout use of tirofiban at
Glycoprotein IIb/IIIa Receptor Blockade 30 days also favored the tirofiban group (26.0% vs. 32.9%,
The platelet glycoprotein IIb/IIIa inhibitors have been the sub- p ¼ 0.020), with longer transfer times being associated with a
ject of extensive clinical trials evaluating their efficacy and greater efficacy advantage. Importantly, tirofiban treatment did
safety in ACS, including a number of randomized controlled not result in a significant increase in major bleeding (4.0% with
studies specific to the STEMI indication. In STEMI, these agents tirofiban vs. 2.9% with placebo, p ¼ 0.363).
have been shown to decrease ischemic complications (62,63). These findings stand in contrast to the results of the
Specifically, a meta-analysis of the 27,115 patients in 11 Harmonizing Outcomes with Revascularization and Stents in
randomized controlled trials of the glycoprotein IIb/IIIa recep- AMI (HORIZONS-AMI) trial (65). The HORIZONS-AMI trial
tor antagonist abciximab in STEMI demonstrated a significant was an open-label, randomized trial of 3602 patients presenting
reduction in short-term (30-day) mortality (2.4% vs. 3.4%, p ¼ with an acute STEMI managed with primary PCI. Patients were
0.047) (63). This improvement in mortality was sustained at the given aspirin and a 600-mg loading dose of clopidogrel, and
6- to 12-month time frame (4.4 vs. 6.2%, p ¼ 0.01) in the patients then randomized to either unfractionated heparin with a gly-
treated with primary PCI; in contradistinction, there was no coprotein IIb/IIIa antagonist or bivalirudin alone (provisional
mortality benefit when abciximab was used to facilitate fibri- glycoprotein IIb/IIIa receptor antagonist treatment was
nolysis. Furthermore, when used to facilitate fibrinolysis, abcix- allowed). The primary end point was a net composite of effi-
imab treatment was associated with an increase in major cacy and safety, including cardiovascular death, reinfarction,
bleeding (5.2% with abciximab vs. 3.1% in controls, p < target vessel revascularization for ischemia, stroke, and major
0.001); conversely, there were no differences in rates of major bleeding within 30 days. The primary end point favored the
bleeding in the primary PCI context (4.7% with abciximab vs. bivalirudin arm, with a 24% relative improvement in net events
[17/6/010/10:7:28] [429–443]
at 30 days (9.2% vs. 12.1%, p ¼ 0.005). The difference was transport the patient directly from home to the cardiac catheter-
derived almost entirely from a decrease in major bleeding ization suite will dramatically reduce first medical contact-
events in the bivalirudin arm (4.9% vs. 8.3%, p ¼ 0.001). How- to-balloon time. As a proportion of patients are emergency
ever, there was a statistically significant 1% absolute increase in department “walk-ins,” all hospitals should develop protocols
stent thrombosis within 24 hours in the bivalirudin group (1.3% for the rapid recognition and triage of the chest pain patient.
vs. 0.3%, p ¼ 0.001), and 66% of patients treated with bivalir- Prehospital emergency medical services, referring hospital
udin also received heparin before randomization. At one year, emergency departments, and primary PCI emergency depart-
while composite rates of ischemic cardiac events were essen- ments should be enabled to directly activate the catheterization
tially the same between the two arms, mortality favored the laboratory without requiring the input of a cardiologist. An
bivalirudin group (3.4% vs. 4.8%, p ¼ 0.03). equitable on-call rotation of the interventional cardiologists in a
Given the available evidence, the ACC/AHA guidelines community should be created that emphasizes coverage of the
list platelet glycoprotein IIb/IIIa receptor blockade as a reason- laboratory by a single physician (regardless of practice align-
able (class IIa) therapeutic adjunct, particularly in patients with ment) rather than patient “ownership.” Once primary PCI has
a high thrombus burden on angiography (Table 33.2) (36). been completed, responsibility for the patient should then be
While a stronger level of evidence exists for abciximab therapy, transitioned promptly from the interventional operator to the
the small molecule antagonists (tirofiban and eptifibatide) are patient’s identified cardiologist.
also recognized as equivalent options. Finally, even with A key role of the STEMI network is to enable the admin-
ONTIME 2, there is but limited evidence supporting use of istration of an antithrombotic “cocktail” early in the course of
this class of agents as a preparatory strategy to facilitate management, preferably by emergency medical services and/
reperfusion or improve outcomes when given upstream prior or the emergency department. This regimen should include
to PCI. antiplatelet and anticoagulant therapies (see the previous sec-
tion “Pharmacology”). A suggested regimen includes 325 mg of
Parenteral Anticoagulants chewable aspirin, a thienopyridine loading dose (600 mg of
Despite the plethora of commercially available and investiga- clopidogrel or 60 mg of prasugrel), and an unfractionated
tional anticoagulants, there have curiously been few studies in heparin bolus (50–70 units/kg, to a maximum of 5000 units).
primary PCI evaluating anticoagulants in a direct comparative
fashion. The ACC/AHA guidelines from 2004, 2007, and 2009
(19,35,36) largely reflect the pharmacological and pharmacody- Assessment and Preparation
namic findings of the elective PCI and ACS literature (Table 33.2). The initial evaluation of the STEMI patient should mirror the
Even HORIZONS-AMI (described above) did not compare the evaluation of the acute trauma patient. An abbreviated Aller-
anticoagulant bivalirudin with another anticoagulant directly, gies, Medications, Past Medical History, Last Meal, and Current
but instead permitted heparin treatment prior to bivalirudin Events (AMPLE) history should be obtained. Specific key
administration and had a comparator arm comprised of unfrac- points should include questions about previous contrast reac-
tionated heparin plus a platelet glycoprotein IIb/IIIa receptor tions, chronic antithrombotic therapy, comorbidities (particu-
blocker. Given the remarkably low overall event rates in both larly renal dysfunction, cerebrovascular disease, and peripheral
arms of HORIZONS-AMI, the critical finding may be the vascular disease), and coronary disease and the details thereof.
overall excellent efficacy and safety of a primary PCI strategy An attempt should be made to assess the ability of the patient
using an approach of aspirin, clopidogrel, and an anticoagu- to take and tolerate dual antiplatelet therapy for a minimum of
one year.
lant, along with the recognition that bivalirudin can be given
safely after an initial bolus of heparin. What does appear clear The key anatomic consideration is to identify the infarct
is that anticoagulants, while requisite for PCI, should be titrated territory from the 12-lead electrocardiogram (67), information
to the degree sufficient to prevent the generation and propa- useful for clinical risk determination and the selection of
gation of thrombus while minimizing their adverse potential to catheters (and sequencing of angiographic imaging) during
accentuate bleeding. the primary PCI procedure. Particular attention should be
directed at evaluating clues of left main or three vessel disease
(68). Evidence of diffuse ischemia should also prompt an eval-
CLINICAL ASPECTS uation for aortic dissection. Computerized interpretations,
STEMI Networks while generally reliable, may miss an acute STEMI, so ques-
Successful best practice management of the STEMI patient is tionable electrocardiograms should promptly be over-read by a
presaged by both cooperative regional planning and local physician and placed in the context of the clinical presentation.
implementation (7,51,52,66). A politic that must frequently be In some cases of posterolateral branch occlusion, the electro-
addressed is the coordination of ambulance service coverage cardiogram may show prominent ST depression and tall
areas and transportation patterns since catchment areas for R-waves in the anterior precordial leads rather than classic ST
primary PCI centers frequently overlap. As ambulance trans- segment elevation. Also, 25% of patients with a NSTEMI will
port times of up to 45 to 60 minutes can still result in initial have an occluded artery at the time of angiography, with 66%
medical contact-to-balloon times of <120 minutes, discussions of these being in the posterolateral distribution (69). This argues
of service coverage areas and referral patterns must include not for a low threshold for referral for cardiac catheterization in the
only the immediate community but also the extended coverage patient with the appropriate presentation but nondiagnostic
region. Emergency medical services should be outfitted with changes on the electrocardiogram.
12-lead electrocardiography systems with computerized anal- The physical examination should focus on signs and
ysis and telemetry; a single vendor for all of the ambulance symptoms of both right-sided and left-sided cardiac compro-
services of a given triage area increases the chance for coordi- mise. This includes an assessment of vital signs (particularly
nated communication. Bypassing nonprimary PCI hospitals to pulse rate, blood pressure, and respiratory rate), oxygenation,
[17/6/010/10:7:28] [429–443]
and adventitial pulmonary and cardiac sounds. The tachycar- While the vast majority of coronary interventions in the pri-
dic and hypotensive patient is at highest risk, especially with an mary PCI setting can be performed with six-French systems,
anterior myocardial infarction. Patients presenting with an exchange for a seven- or eight-French guide catheter should be
inferior myocardial infarction and hypotension or who develop considered in the unusual circumstance that diagnostic angiog-
profound hypotension after administration of nitroglycerin raphy identifies a true bifurcation culprit lesion.
should be aggressively resuscitated with crystalloid fluid For the intervention itself, a basic set of interventional
expansion. Right ventricular involvement may substantially tools should be prepared by the scrub assistant at the same time
impair forward cardiac output, resulting in profound hypoten- the diagnostic images are being obtained. These include a soft-
sion even with relatively clear lungs. Left ventricular cardio- tipped 0.014-in coronary guidewire, a thrombus aspiration
genic shock, on the other hand, frequently presents with catheter, and a 2.0- to 2.5-mm coronary balloon sized to be at
hypotension and pulmonary edema. A rapid assessment least 0.25 to 0.5 mm smaller than the nominal vessel diameter. If
should be made as to whether an intra-aortic balloon pump a platelet glycoprotein IIb/IIIa receptor antagonist is to be
(or other form of circulatory support), transvenous pacemaker, administered, the bolus and infusion should also be initiated
and/or intubation are immediately needed to manage hypo- at the same time. A coil-tipped coronary guidewire (rather than
tension, bradyarrhythmias, and/or ventilatory compromise, a polymer-coated hydrophilic guidewire) is generally preferred
respectively. It is recommended that these procedures be to decrease the potential for coronary dissection and perfora-
completed prior to the initiation of coronary arteriography, tion. Should a hydrophilic guidewire ultimately be required,
even if this prolongs the ultimate DB time, as contrast injection successful placement of the wire in the distal lumen should be
and PCI may result in further hemodynamic decompensation. confirmed before intervention is performed, even if this
requires removal of the guidewire and injection of contrast
through a (nonrapid exchange) balloon catheter or an end-hole
TECHNICAL FUNDAMENTALS infusion catheter. Our preference is to start with a 300-cm
Procedure and Technique guidewire as this allows the balloon to be preloaded, provides
Upon arrival in the laboratory, monitoring and/or defibrilla- increased support for advancing interventional devices across
tion patches should be placed on the patient and both groins the infarct lesion, and facilitates the injection of contrast and
prepared. Once the patient has been prepped and draped, the medications through the balloon lumen into the distal artery.
first technical task is establishing vascular access. By this point, Crossing the lesion with a guidewire may at least par-
the patient should have been treated with the antithrombotic tially restore epicardial blood flow. In STEMI involving a right
cocktail (or variant, depending on local protocol) described coronary or dominant left circumflex artery, reperfusion may
above. When using the femoral approach, fluoroscopy (rather precipitate the vagally mediated Bezold-Jarish reflex (71).
than the inguinal crease) should be employed to identify ana- Administration of 0.5 mg of atropine before the reestablishment
tomic landmarks, particularly the location of the head of the of flow, especially when a narrow pulse pressure or relative
femur. The femoral artery should be accessed with a single bradycardia are present, should at least partially mitigate this
anterior wall puncture, with the target zone being between the reflex. The reflex nominally lasts 5 to 10 minutes; if longer than
lower margin and mid portion of the femoral head. Transradial 15 minutes, other etiologies of hypotension should be consid-
access has been used quite successfully in primary PCI (70); ered. If significant disease exists in the noninfarct related
however, the operator should first become facile with this arteries, the development of the Bezold-Jarisch reflex with
approach via a series of elective cases before attempting to severe hypotension can precipitate cardiogenic shock or closure
perform primary PCI via the radial artery. of a noninfarct artery.
Before the first coronary arteriogram, a quick reassess- The obvious goal of primary PCI is to restore normal
ment should be performed to evaluate the cardiopulmonary epicardial coronary blood flow and myocardial microvascular
status of the patient. Placement of an intra-aortic balloon at this perfusion. Table 33.2 lists the most commonly used grading
juncture should be considered if the systolic blood pressure is scales related to these concepts: thrombolysis in myocardial
90 mm or less or if the patient has developed pulmonary edema infarction (TIMI) flow and TIMI myocardial perfusion grade
(Killip class III or IV). It should be anticipated that coronary (56,57). Another measure, the corrected TIMI frame count, has
arteriography and intervention will transiently impair cardiac the advantage of being a continuous, quantifiable, and repro-
performance, and the contrast load can further compromise ducible measure of coronary perfusion (72). Achievement of
oxygenation. Even though placement of an intra-aortic balloon TIMI 3 flow remains a most powerful predictor of mortality at
pump necessitates access via a second vascular access site, 30 days and one year, with restoration of normal myocardial
stability of the patient through the procedure and beyond perfusion having perhaps an even stronger predictive value.
should be the paramount concern. Once the coronary guidewire has been advanced into the
The initial set of coronary arteriograms should evaluate distal true lumen, the first mode of intervention must be
the (anticipated) noninfarct coronary artery. Typically, a stan- selected. The principal choices are balloon angioplasty, direct
dard Judkins diagnostic catheter serves this purpose well. The stenting (without predilation with a balloon catheter), or aspi-
caution is that the noninfarct artery may supply substantial ration thrombectomy. The Thrombus Aspiration During Percu-
collateral supply, with angiography precipitating ischemia. A taneous Coronary Intervention in Acute Myocardial Infarction
brief delay of a few additional seconds relative to the normal Study (TAPAS), which randomized 1071 primary PCI patients
pace of coronary injections will allow time for myocardial to either thrombus aspiration with the Export Aspiration Cath-
recovery from contrast-induced ischemia. Once angiography eter (Medtronic, Minneapolis, Minnesota, U.S.) or conventional
of the noninfarct related has been completed, a guide catheter treatment, documented a 46% relative reduction in cardiac
(rather than diagnostic catheter) should be advanced to the mortality at one year with aspiration (6.7% vs. 3.6% with
contralateral coronary artery to perform diagnostic arteriogra- aspiration, p ¼ 0.020) and a 43% reduction in death or nonfatal
phy and obtain diagnostic images of the infarct-related lesion. myocardial infarction (9.9% vs. 5.6% with aspiration, p ¼ 0.009)
[17/6/010/10:7:28] [429–443]
Figure 33.5 (A) Kaplan-Meier survival curves of

1815 patients in seven trials of aspiration “manual”
thrombectomy [Export (Medtronic, Minneapolis,
Minnesota, U.S.), Pronto (Vascular Solutions, Min-
neapolis, Minnesota, U.S.), and Diver CE (Invatec,
Roncadelle, Brescia, Italy) catheters] versus stan-
dard PCI controls (log rank p = 0.011). (B) Kaplan-
Meier survival curves of 871 patients in four trials
of device-based “nonmanual” thrombectomy
[AngioJet (Medrad Interventional/Possis, Minne-
apolis, Minnesota, U.S.), X-Sizer (ev3 Endovascu-
lar, Plymouth, Minnesota, U.S.), TVAC (Nipro,
Osaka, Japan), and Rescue PT systems (Boston
Scientific, Maple Grove, Minnesota, U.S.)] versus
standard PCI controls (log rank p ¼ 0.481).
(73). After primary PCI, microvascular perfusion improved to Diver CE (Invatec, Roncadelle, Brescia, Italy) aspiration catheters
final TIMI myocardial blush grade 2 or 3 in 73.7% in the con- (Fig. 33.5) (76). In aggregate, these data suggest that manual
ventional treatment group and 82.9% in the aspiration group (p < aspiration thrombectomy should be considered the first interven-
0.001), with histopathological examination documenting success- tional option where technically feasible.
ful thrombus aspiration in 72.9% of patients (74). A meta-analysis If balloon angioplasty is chosen as the initial intervention,
of 21 trials of a variety of thrombectomy devices available before the goal should generally not be to achieve a definitive result
2007 (preceding the TAPAS trial) had previously demonstrated with balloon angioplasty alone, unless a stent cannot be deliv-
an improvement in myocardial blush scores (but no mortality ered to the infarct lesion or a clinical consideration (such as
benefit); (75) an updated analysis has subsequently demonstrated impending surgery) dictates otherwise. Compression of soft
an all-cause mortality benefit at a median of 365 days in 11 thrombus can elaborate vasoactive substances and cause distal
randomized trials of thrombectomy versus PCI without throm- embolization of thromboembolic particulate material, and
bectomy (p ¼ 0.049), with the benefit confined to patients treated aggressive balloon dilation can produce arterial dissection.
in the manual (aspiration) thrombectomy trials of the Export, Instead, the goals of balloon angioplasty should be to restore
Pronto (Vascular Solutions, Minneapolis, Minnesota, U.S.), and epicardial flow and to prepare the artery for stent implantation.
[17/6/010/10:7:28] [429–443]
The current primary PCI approach anticipates the Support of the Critically Ill Patient
implantation of a coronary stent as the definitive intervention. In the hemodynamically unstable STEMI patient, changes to
Interestingly, this approach was still the topic of debate into the above procedural sequence may need to be considered.
the early 2000s (77). In the five-year follow-up of the Primary While ventriculography can usually be deferred until after PCI
Angioplasty in Myocardial Infarction (PAMI) trial (n ¼ 2087, has been completed (to reduce DB time), ventriculography
stents in 692), patients who underwent stent implantation (not should be performed prior to PCI in the unstable patient.
as a randomized treatment) experienced a lower mortality This is to exclude mechanical complications of STEMI such as
than those treated solely with balloon angioplasty (78). How- flail mitral leaflet, ventricular septal rupture, pseudoaneurysm,
ever, randomized comparisons, notably the Stent Primary and free wall rupture. An ascending aorta aortogram should be
Angioplasty in Myocardial Infarction (Stent PAMI) study ini- performed in the patient with diffuse ST segment depression
tially were inconsistent with this finding. In Stent PAMI, 900 (or diffuse elevation) on the electrocardiogram to evaluate for
patients were randomized to balloon angioplasty or implan- aortic dissection. The presence of any of these conditions
tation of the first generation Palmaz-Schatz (Cordis Corpora- should be considered a contraindication to PCI and should
tion, Miami Lakes, Florida, U.S.) stent (79). While patients prompt emergency cardiothoracic surgical consultation.
assigned to balloon angioplasty had higher rates of restenosis, Equipment in the catheterization laboratory must antici-
target vessel revascularization, and more angina at six months pate the need to provide cardiopulmonary support in the
compared with the stent arm, stent implantation was associ- critically ill patient. Supplies that must be immediately avail-
ated with a strong trend toward increased mortality, raising able include a cardiopulmonary resuscitation cart, intra-aortic
concerns about stent implantation in STEMI. These issues balloon system, and external and transvenous temporary pace-
informed the design and execution of the Controlled Abcix- maker supplies. The threshold for intubation should be quite
imab and Device Investigation to Lower Late Angioplasty low, not only for obvious pulmonary embarrassment, but also
Complications (CADILLAC) study, a trial with a 2 2 facto- in the delirious or otherwise uncooperative patient; an inability
rial design that evaluated both a second generation (Multi- to cooperate or follow instructions is often an early sign of
Link, Guidant Corporation, Mountain View, California, U.S.) respiratory embarassment.
stent versus balloon angioplasty and treatment with the As highlighted above, patients in frank cardiogenic shock
platelet glycoprotein IIb/IIIa receptor blocker abciximab ver- should undergo immediate intra-aortic balloon pump implan-
sus no treatment (62,80). The key findings of CADILLAC were tation before the initiation of coronary arteriography. Place-
a reduction in restenosis at one year with stent implantation ment of an intra-aortic balloon pump in this setting has been
compared with balloon angioplasty (without a difference in demonstrated to improve short-term outcomes (82) and should
mortality) and a reduction in recurrent myocardial infarction not require more than three to five minutes in experienced
and infarct-related artery thrombosis with abciximab treat- hands. Of note, however, the prospective randomized PAMI-II
ment. A number of additional trials have since been con- trial of intra-aortic balloon counterpulsation in 437 high-risk
ducted with stent implantation in primary PCI, confirming patients, (out of a cohort of 1100 acute STEMI patients), pro-
the value of this strategy in maintaining arterial patency and phylactic placement of an intra-aortic balloon pump did not
improving short- and long-term outcomes (46,53). improve in-hospital outcomes (83).
In STEMI, some degree of vasoconstriction is more often Multivessel disease identified at the time of STEMI
the rule than the exception. This can lead to an underestimation remains a challenging scenario. In the absence of cardiogenic
of the nominal reference vessel diameter. Small amounts of shock, identification and treatment of the culprit lesion with
intracoronary vasodilators (nitroglycerin or nitroprusside) as deferral of additional (nonculprit) PCI remains the guideline
tolerated by the patient’s blood pressure should therefore be (19,35,36). Unsuccessful or otherwise compromised nonculprit
given prior to stent implantation to aid in sizing the stent. A vessel PCI can have unpredictable and even disastrous con-
general rule of thumb is to undersize the diameter of the initial sequences in the acute STEMI patient. Multilesion, multivessel
dilation balloon by 0.25 to 0.5 mm, while the opposite is true PCI at the time of the initial primary PCI remains to be
for stent selection. A slightly longer (5–10 mm) and larger systematically evaluated in large-scale trials.
(0.25–0.5 mm) stent should be selected for implantation in the
infarct target lesion. Mastery of the compliance curves of stent
systems is thus a prerequisite for the primary PCI operator. The Complications
goal should be to deploy the stent at a high pressure without In the STEMI setting, the optimal management of thrombus
further postdilation, or a “one and done” philosophy. Additional includes dissolution and/or aspiration. However, even with
dilations of an implanted stent can precipitate microemboliza- optimal pharmacotherapy and thrombus aspiration, emboliza-
tion and potentiate the no-reflow phenomenon. Once epicardial tion may occur and lead to compromise of the distal coronary
flow has been reestablished, the primary determinant of myo- bed (84). Large emboli in the distal epicardial coronary can
cardial recovery will be myocardial perfusion, not epicardial often be disrupted sufficiently to restore flow by advancing an
flow. The failure of at least 50% ST segment elevation resolution interventional device (e.g., guidewire, balloon, aspiration cath-
from the initial ECG after reperfusion is a strong predictor of one eter) through the thrombus. Selective intracoronary infusion of
year mortality (81). Every effort should therefore be made to platelet glycoprotein IIb/IIIa receptor inhibitors can also
avoid further compromise of the microcirculation. Direct stent- restore perfusion.
ing, while having the potential to reduce distal embolization, At the microvascular level, the no-reflow phenomenon
remains to be rigorously evaluated as an independent variable in can complicate otherwise successful reperfusion of the epicar-
primary PCI. Because of the potential for size and length mis- dial artery. This phenomenon is characterized angiographically
match of the stent to the lesion and the nominal vessel diameter, by a poor myocardial blush and reflects inadequate myocellular
particularly when there is TIMI 0 or 1 flow, we generally dis- perfusion. It occurs in 10% to 40% of acute STEMI patients,
courage direct stenting as the first intervention in primary PCI. frequently complicating primary PCI (84–87). The key to
[17/6/010/10:7:28] [429–443]
management is prevention; to reduce the probability of no-

reflow, “light-touch” techniques similar to those used in saphe-
nous vein graft PCI should be employed in the treatment of
infarct lesions. When it does occur, no-reflow can produce
myocardial ischemia, worsen myocardial necrosis, extend myo-
cardial infarction, cause ventricular arrhythmias, and result in
rapid hemodynamic deterioration. The occurrence of no-reflow
demands a rapid response. Intracoronary administration of a
calcium channel blocker, adenosine, or sodium nitroprusside
through an end-hole infusion catheter positioned beyond the
target lesion may help resolve no-reflow. Patients who develop
persistent ST elevation have a significantly higher mortality.
Fortunately, even without demonstrable improvement in left
ventricular contractile function, reversal of no-reflow can have
favorable effects on remodeling of the left ventricle (88). The
most important element of the differential is major coronary
artery dissection. If further stent implantation is elected, it is
absolutely critical to confirm that the coronary guidewire is in
the true lumen distally. Finally, although it has become routine Figure 33.6 Results of a meta-analysis demonstrating reductions
in both 30-day mortality and the composite of death and reinfarction
to administer vasodilators to help resolve no-reflow, large
in five trials of emergency rescue PCI for failed fibrinolysis versus
prospective randomized trials are lacking.
conservative management. Abbreviation: PCI, percutaneous coro-
nary intervention. Source: Adapted from Ref. 94.
PCI AFTER FIBRINOLYSIS
While the focus of this chapter has been to develop, describe
and delineate the primary PCI strategy in acute STEMI, there
nonetheless remain a substantial proportion of patients who
will be initially managed with fibrinolytic therapy. Several death or reinfarction compared with a conservative approach
constructs therefore deserve mention in the context of an (Fig. 33.6) (94,95).
intervention-based approach to the patient with STEMI. Rescue PCI (rather than repeat fibrinolysis or conserva-
tive management) has thus emerged as the preferred strategy
for patients without evidence of successful reperfusion 60 to
Rescue PCI 90 minutes after fibrinolytic administration. This is especially
Failure of fibrinolysis to restore perfusion within 60 to 90 minutes true for patients with a large amount of myocardium at risk
of administration is associated with reduced left ventricular and hemodynamic or electrical instability. Bleeding risk should
function and increased rates of mechanical complications and be taken into consideration as the pooled analysis did demon-
mortality compared with successful reperfusion (89–91). The strate a significant increase in the combined end point of major
clinical markers of reperfusion, such as chest pain diminution, and minor bleeding (11.9%) in the rescue PCI arm compared with
partial resolution of ST segments (>50–70%) and the occurrence the conservative arm (1.3%, p < 0.001). In particular, bleeding
of reperfusion arrhythmias unfortunately are of limited pre- events were associated with concomitant abciximab therapy and
dictive value (92). Califf and colleagues studied 386 patients were typically associated with the arterial access site.
undergoing angiography 90 minutes after the initiation of tPA
for STEMI and found that while 96% of patients who had
complete resolution of ST segment elevation had coronary Routine Early PCI After Fibrinolysis
patency, this finding was relatively rare as it was seen in only A variant of rescue PCI for failed fibrinolysis is routine early
6% of patients (93). Among patients with partial improvement in catheterization following (presumably successful) fibrinolysis
ST segment elevation, 84% had coronary patency, but even (94,95). In the Collet meta-analysis described above, this
this finding occurred in only 38% of patients. Unchanged ST approach was also analyzed, specifically differentiating “bal-
segment elevation was associated with a 63% patency rate and loon era” from “stent era” trials. Studies included in the
unchanged or worsened chest pain was associated with a analysis employed a strategy of systematic and early (<24
60% patency rate. The presence or absence of arrhythmias also hours) PCI versus delayed or ischemia-guided PCI. The authors
did not appear to be closely associated with coronary patency. Of found that the three stent era studies of early catheterization
the patients without either ST segment or symptom resolution, with PCI demonstrated a strong trend toward reduction in
56% were still found to have a patent infarct artery at 90 minutes. mortality compared with a delayed or ischemia-guided PCI
Given these uncertainties, the approach of urgent transfer approach (3.8% vs. 6.7%) (OR 0.56, 95% CI 0.29–1.05, p ¼ 0.07)
following fibrinolysis for immediate catheterization and rescue and a twofold reduction in the rate of death or reinfarction
PCI has been espoused as a strategy to increase reperfusion, (7.5% vs. 13.2%) (OR 0.53, 95% CI 0.33–0.83, p ¼ 0.0067). Since
prevent further myonecrosis, and improve outcomes. There are then, the large-scale 1059-patient Trial of Routine Angioplasty
currently five published randomized trials that have enrolled and Stenting After Fibrinolysis to Enhance Reperfusion in
920 patients evaluating this approach. In a meta-analysis of Acute Myocardial Infarction (TRANSFER-AMI) has confirmed
these trials performed by Collet and colleagues, the immediate the advantage of early referral for cardiac catheterization and
catheterization with rescue PCI approach was associated with a PCI following fibrinolysis (Table 33.3) (96). Of note, while the
35% relative risk reduction (p ¼ 0.04) in mortality and a 36% ACC/AHA guidelines list routine cardiac catheterization after
relative risk reduction (p ¼ 0.009) in the combined end point of fibrinolysis as class IIa in high risk patients and IIb in low risk
440
[17/6/010/10:7:28] [429–443]
Table 33.3 Trials Evaluating a Strategy of Systematic, Early Cardiac Catheterization with Intended PCI After Clinically Successful Fibrinolysis Versus Delayed or Ischemia-
Guided Catheterization
End point event rate
Symptom Primary End point Routine Early
Trial Year n onset Fibrinolytic end point follow-up (%) PCI (%) p
Southwest German Interventional 2003 197 <12 hr rPA D/RI/TVR/ 6 mo 50.6 25.6 0.001
Study in Acute Myocardial ischemia
Infarction III (SIAM III)
Grupo de Análisis de la Cardiopatı́a 2004 500 <12 hr Tissue D/RI/TVR 12 mo 20.3 9.3 008
Isquémica Aguda 1 plasminogen
(GRACIA-1) activator
Combined Angioplasty and 2005 170 <6 hr TNK D/RI/ischemia/ 6 mo 24.4 11.6 0.04
Pharmacological Intervention stroke

Versus Thrombolysis Alone
CARDIOVASCULAR CATHETERIZATION AND INTERVENTION
in Acute Myocardial Infarction

(CAPITAL AMI)
Combined Abciximab Reteplase 2009 600 <12 hr Half-dose rPA D/RI/ischemia 30 days 10.7 4.4 0.004
Stent Study in Acute Myocardial
Infarction (CARESS-in-AMI)
Trial of Routine Angioplasty and 2009 1059 <12 hr TNK D/RI/ischemia/ 30 days 17.2 11.0 0.004
Stenting After Fibrinolysis to congestive
Enhance Reperfusion in Acute heart failure/
Myocardial Infarction cardiogenic
(TRANSFER-AMI) shock
Abbreviations: PCI, percutaneous coronary intervention; TNK, tenecteplase; rPA, reteplase; D, death; RI, reinfarction; TVR, target vessel revascularization.
Source: From Refs. 96–100.
[17/6/010/10:7:28] [429–443]
patients, the European Society of Cardiology guidelines assign acute myocardial infarction. Anglo-Scandinavian Study of Early
this a class I status (101,102). While outcomes in the balloon Thrombolysis (ASSET). Lancet 1988; 2:525–530.
era were indeed worse with coronary intervention, improve- 15. Rossi P, Bolognese L on behalf of Urochinasi per via Sistemica
ments in technique and technologies of the stent era have nell’Infarto Miocardico (USIM) Collaborative Group. Compari-
son of intravenous urokinase plus heparin versus heparin alone
dramatically reversed this finding in favor of early and sys-
in acute myocardial infarction. Am J Cardiol 1991; 68:585–592.
tematic cardiac catheterization with immediate percutaneous 16. ISIS-3 (Third International Study of Infarct Survival) Collabora-
revascularization. tive Group. ISIS-3: a randomized trial of streptokinase vs. tissue
In summary, a percutaneous revascularization strategy plasminogen activator vs. anistreplase and of aspirin plus hep-
focused on restoring (and maintaining) patency has indeed arin vs. aspirin alone among 41,299 cases of suspected acute
become the recommended approach for managing the patient myocardial infarction. Lancet 1992; 339:753–770.
presenting with STEMI, whether via primary PCI or as an 17. EMERAS (Estudio Multicentrico Estreptoquinasa Repúblicas de
adjunct to augment fibrinolysis. Finally, regardless of initial América del Sur) Collaborative Group. Randomised trial of late
strategy, the need for advanced PCI capabilities including thrombolysis in patients with suspected acute myocardial infarc-
cardiopulmonary support also argues for the transfer of all tion. Lancet 1993; 342:767–772.
18. LATE Study Group. Late Assessment of Thrombolytic Efficacy
STEMI patients to tertiary centers of excellence as we collec-
(LATE) study with alteplase 6–24 hours after onset of acute
tively strive to improve clinical outcomes. myocardial infarction. Lancet 1993; 342:759–766.
19. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guide-
lines for the management of patients with ST-elevation myocar-
REFERENCES dial infarction—executive summary. A report of the American
1. Falk E, Shah PK, Fuster V. Coronary plaque disruption. Circula- College of Cardiology/American Heart Association Task Force
tion 1995; 92:657–671. on Practice Guidelines (Writing Committee to revise the 1999
2. Virmani R, Burke AP, Kolodgie FD, et al. Pathology of the thin- guidelines for the management of patients with acute myocardial
cap fibroatheroma: a type of vulnerable plaque. J Interv Cardiol infarction). J Am Coll Cardiol 2004; 44:671–719.
2003; 16:267–272. 20. Gore JM, Granger CB, Simoons ML, et al. Stroke after thrombol-
3. Fuster V, Badimon L, Badimon J, et al. The pathogenesis of ysis. Mortality and functional outcomes in the GUSTO-I trial.
coronary artery disease and the acute coronary syndromes. N Global Use of Strategies to Open Occluded Coronary Arteries.
Engl J Med 1992; 326:242–250. Circulation 1995; 92:2811–2818.
4. Davies M, Thomas A. Plaque fissuring-the cause of acute myo- 21. Van de Werf F, Adgey J, Ardissino D, et al. Single-bolus ten-
cardial infarction, sudden ischemic death and crescendo angina. ecteplase compared with front-loaded alteplase in acute myocar-
Br Heart J 1985; 53:363–373. dial infarction: the ASSENT-2 double-blind randomized trial.
5. DeWood MA, Spores J, Notske R, et al. Prevalence of total Lancet 1999; 354:716–722.
coronary occlusion during the early hours of transmural myo- 22. The Global Use of Strategies to Open Occluded Coronary
cardial infarction. N Engl J Med 1980; 303:897–902. Arteries (GUSTO III) Investigators. A comparison of reteplase
6. Ryan T, Anderson J, Antman E, et al. ACC/AHA guidelines for with alteplase for acute myocardial infarction. N Engl J Med
the management of patients with acute myocardial infarction. A 1997; 337:1118–1123.
report of the American College of Cardiology/American Heart 23. Gurwitz JH, Gore JM, Goldberg RJ, et al. Risk for intracranial
Association Task Force on Practice Guidelines (Committee on hemorrhage after tissue plasminogen activator treatment for
Management of Acute Myocardial Infarction). J Am Coll Cardiol acute myocardial infarction. Ann Intern Med 1998; 129:597–604.
1996; 28:1328–1428. 24. Berkowitz SD, Granger CB, Pieper KS, et al. Incidence and
7. Tcheng J. Primary Angioplasty in Acute Myocardial Infarction. predictors of bleeding after contemporary thrombolytic therapy
2nd ed. New York: Humana Press. 2009. for myocardial infarction. The Global Utilization of Streptokinase
8. Rentrop K, Blanke H, Karsch K, et al. Acute myocardial infarc- and Tissue Plasminogen activator for Occluded coronary arteries
tion: intracoronary application of nitroglycerin and streptoki- (GUSTO) I Investigators. Circulation 1997; 95:2508–2516.
nase. Clin Cardiol 1979; 2:354–363. 25. Lincoff AM, Topol EJ. Illusion of reperfusion. Does anyone
9. Fibrinolytic Therapy Trialists (FTT) Collaborative Group. Indica- achieve optimal reperfusion during acute myocardial infarction?
tions for fibrinolytic therapy in suspected acute myocardial Circulation 1993; 88:1361–1374.
infarction: collaborative overview of early mortality and major 26. Mueller RL, Sanborn TA. The history of interventional cardiol-
morbidity results from all randomized trials of more than 1000 ogy: cardiac catheterization, angioplasty, and related interven-
patients. Lancet 1994; 343:311–322. tions. Am Heart J 1995; 129:146–172.
10. GISSI (Gruppo Italiano per lo Studio della Streptochinasi nell’In- 27. Gruntzig A. Transluminal dilatation of coronary-artery stenosis.
farto miocardico). Effectiveness of intravenous thrombolytic Lancet 1978; 1:263.
treatment in acute myocardial infarction. Lancet 1986; 1:397–401. 28. Rentrop P, De Vivie ER, Karsch KR, et al. Acute coronary occlusion
11. ISAM (Intravenous Streptokinase in Acute Myocardial Infarc- with impending infarction as an angiographic complication relieved
tion) Study Group. A prospective trial of intravenous streptoki- by a guide-wire recanalization. Clin Cardiol 1978; 1:101–106.
nase in acute myocardial infarction (ISAM): mortality, morbidity, 29. Geoffrey Hartzler interview, available at http://www.ptca.org/
and infarct size at 21 days. N Engl J Med 1986; 314:1465–1471. history.html. Accessed July 1, 2009.
12. AIMS (APSAC Intervention Mortality Study) Trial Study Group. 30. Adams J. I’ll Get in Shape Tomorrow . . . Durham Morning
Effects of intravenous APSAC on mortality after acute myocar- Herald, Durham, NC, 1981:D1.
dial infarction: preliminary report of a placebo-controlled clinical 31. Grines CL, Browne KF, Marco J, et al. A comparison of immedi-
trial. Lancet 1988; 331:545–549. ate angioplasty with thrombolytic therapy for acute myocardial
13. ISIS-2 (Second International Study of Infarct Survival) Collabo- infarction. The Primary Angioplasty in Myocardial Infarction
rative Group. Randomised trial of intravenous streptokinase, oral Study Group. N Engl J Med 1993; 328:673–679.
aspirin, both, or neither among 17,187 cases of suspected acute 32. Zijlstra F, de Boer MJ, Hoorntje JCA, et al. A comparison of
myocardial infarction: ISIS-2. Lancet 1988; 2:349–360. immediate coronary angioplasty with intravenous streptokinase
14. Wilcox G, von der Lippe CG, Olsson G, et al. for the ASSET in acute myocardial infarction. N Engl J Med 1993; 328:680–684.
(Anglo-Scandinavian Study of Early Thrombolysis) Study Group. 33. Gibbons RJ, Holmes DR, Reeder GS, et al. Immediate angioplasty
Trial of tissue plasminogen activator for mortality reduction in compared with the administration of a thrombolytic agent
[17/6/010/10:7:28] [429–443]
followed by conservative treatment for myocardial infarction. N pooled analysis of randomized clinical trials comparing primary
Engl J Med 1993; 328:685–691. percutaneous coronary intervention and in-hospital fibrinolysis in
34. Keeley EC, Boura JA, Grines CL. Primary angioplasty vs. intrave- acute myocardial infarction patients. Eur Heart J 2006; 27:779–788.
nous thrombolytic therapy for acute myocardial infarction, a quan- 51. Henry TD, Atkins JM, Cunningham MS, et al. ST-segment ele-
titative review of 23 randomized trials. Lancet 2003; 361:13–20. vation myocardial infarction: recommendations on triage of
35. Antman EM, Hand M, Armstrong PW, et al. 2007 focused update patients to heart attack centers: Is it time for a national policy
of the ACC/AHA 2004 guidelines for the management of patients for the treatment of ST-segment elevation myocardial infarction?
with ST-elevation myocardial infarction: a report of the American J Am Coll Cardiol 2006; 47:1339–1345.
College of Cardiology/American Heart Association Task Force on 52. Jollis JG, Roettig ML, Aluko AO, et al, for the Reperfusion of
Practice Guidelines. J Am Coll Cardiol 2008; 51:210–247. Acute Myocardial Infarction in North Carolina Emergency
36. Kushner FG, Hand M, Smith SC Jr, et al. 2009 focused updates: Departments (RACE) Investigators. Implementation of a state-
ACC/AHA guidelines for the management of patients with wide system for coronary reperfusion for ST-segment elevation
ST-elevation myocardial infarction (updating the 2004 guideline myocardial infarction. JAMA 2007; 298:2371–2380.
and 2007 focused update) and ACC/AHA/SCAI guidelines on 53. Bates ER, Nallamothu BK. Commentary: the role of percutaneous
percutaneous coronary intervention (updating the 2005 guideline coronary intervention in ST-segment-elevation myocardial
and 2007 focused update): a report of the American College of infarction. Circulation 2008; 188:567–573.
Cardiology Foundation/American Heart Association Task Force 54. Sabatine MS, Cannon CP, Gibson CM, et al. Addition of clopi-
on Practice Guidelines. J Am Coll Cardiol 2009; 54:2205–2241. dogrel to aspirin and fibrinolytic therapy for myocardial infarction
37. Brener SJ. Catheter-based reperfusion for acute myocardial infarc- with ST-segment elevation. N Engl J Med 2005; 352:1179–1189.
tion. In: Topol E, ed. Textbook of Interventional Cardiology. 5th 55. Sabatine MS, Cannon CP, Gibson CM, et al. Effect of clopidogrel
ed. Philadelphia: WB Saunders. 2008:285–302. pretreatment before percutaneous coronary intervention in
38. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization patients with ST-elevation myocardial infarction treated with
in acute myocardial infarction complicated by cardiogenic shock: fibrinolytics: the PCI-CLARITY study. JAMA 2005; 294:1224–1232.
SHOCK Investigators. Should we emergently revascularize 56. The TIMI Study Group. The thrombolysis in myocardial infarc-
occluded coronaries for cardiogenic shock? N Engl J Med 1999; tion (TIMI) trial. N Engl J Med 1985; 312:932–936.
341:625–634. 57. Gibson CM, Cannon CP, Murphy SA, et al. Relationship of TIMI
39. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization Myocardial Perfusion grade to mortality after administration of
and long-term survival in cardiogenic shock complicating acute thrombolytic drugs. Circulation 2000; 101:125–130.
myocardial infarction. JAMA 2006; 295:2511–2515. 58. Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versus
40. Dzavik V, Sleeper LA, Picard MH, et al. for the SHOCK Inves- clopiodogrel in patients with acute coronary syndromes. N Engl J
tigators. Outcome of patients aged 75 years in the SHould we Med 2007; 357:2001–2015.
emergently revascularize Occluded Coronaries in cardiogenic 59. Montalescot G, Wiviott SD, Braunwald E, et al. for the TRITON-
shocK (SHOCK) trial: Do elderly patients with acute myocardial TIMI 38 Investigators. Prasugrel compared with clopidogrel in
infarction complicated by cardiogenic shock respond differently patients undergoing percutaneous coronary intervention for
to emergent revascularization? Am Heart J 2005; 149:1128–1134. ST-elevation myocardial infarction (TRITON-TIMI 38): double-
41. Grines C. Senior PAMI: a prospective randomized trial of pri- blind, randomized controlled trial. Lancet 2009; 373:723–731.
mary angioplasty and thrombolytic therapy in elderly patients 60. Wiviott SD, Antman EM, Gibson CM, et al. Evaluation of
with acute myocardial infarction. Presented at: Transcatheter prasugrel compared with clopidogrel in patients with acute
Therapeutics (TCT). Washington, D.C., 2005. Available at: coronary syndromes: design and rationale for the Trial to assess
http://www.theheart.org/article/581549.do. Improvement in Therapeutic Outcomes by optimizing platelet
42. Andersen HR, Nielsen TT, Rasmussen K, et al. A comparison of InhibitioN with prasuegrel Thrombolysis in Myocardial Infarc-
coronary angioplasty with fibrinolytic therapy in acute myocar- tion 38 (TRITON-TIMI38). Am Heart J 2006; 152:627–635.
dial infarction. N Engl J Med 2003; 349:733–742. 61. Brandt JT, Payne CD, Wiviott SD, et al. A comparison of
43. Widimsky P, Budesinsky T, Vorác D, et al. Long distance trans- prasugrel and clopidogrel loading doses on platelet function:
port for primary angioplasty vs. immediate thrombolysis in acute magnitude of platelet inhibition is related to active metabolite
myocardial infarction: final results of the randomized national formation. Am Heart J 2007; 153:e9–e16.
multicentre trial—PRAGUE-2. Eur Heart J 2003; 24:94–104. 62. Tcheng JE, Kandzari DE, Grines CL, et al. Benefits and risks of
44. Nallamothu BK, Bates ER, Herrin J. Times-to-treatment in trans- abciximab use in primary angioplasty for acute myocardial
fer patients undergoing primary percutaneous coronary inter- infarction: the Controlled Abciximab and Device Investigation
vention in the United States: a National Registry of Myocardial to Lower Late Angioplasty Complications (CADILLAC) trial.
Infarction-3/4 analysis. Circulation. 2005; 111:761–767. Circulation 2003; 108:1316–1323.
45. Dalby M, Bouzamondo A, Lechat PH, et al. Transfer for primary 63. De Luca G, Suryapranata H, Stone GW, et al. Abciximab as
angioplasty versus immediate thrombolysis in acute myocardial adjunctive therapy to reperfusion in acute ST-segment elevation
infarction: a meta-analysis. Circulation 2003; 108:1809–1814. myocardial infarction: a meta-analysis of randomized trials.
46. Stone GW. Angioplasty strategies in ST-segment elevation myo- JAMA 2005; 293:1759–1765.
cardial infarction: part I: primary percutaneous coronary inter- 64. Van’t Hof AWJ, Ten Berg J, Heestermans T, et al. Prehospital
vention. Circulation 2008; 118:538–551. initiation of tirofiban in patients with ST-elevation myocardial
47. Nallamothu BK, Bates ER. Percutaneous coronary intervention infarction undergoing primary angioplasty (On-TIME 2): a multi-
versus fibrinolytic therapy in acute myocardial infarction: Is centre, double-blind, randomized controlled trial. Lancet 2008;
timing (almost) everything? Am J Cardiol 2003; 92:824–826. 372:537–546.
48. Betriu A, Masotti M. Comparison of mortality rates in acute 65. Stone GW, Witzenbichler B, Guagliumi G, et al. Bivalirudin
myocardial infarction treated by percutaneous coronary inter- during primary PCI in acute myocardial infarction. N Engl J
vention versus fibrinolysis. Am J Cardiol 2005; 95:100–101. Med 2008; 358:2218–2230.
49. Pinto DS, Kirtane AJ, Nallamothu BK, et al. Hospital delays in 66. Kalla K, Christ G, Karnik R, et al. Implementation of guidelines
reperfusion for ST-elevation myocardial infarction: implications improves the standard of care: the Viennese registry on reperfu-
when selecting a reperfusion strategy. Circulation 2006; 114: sion strategies in ST-elevation myocardial infarction (Vienna
2019–2025. STEMI registry). Circulation 2006; 113:2398–2405.
50. Boersma E. The Primary Coronary Angioplasty vs. Thrombolysis 67. Zimetbaum PJ, Josephson ME. Use of the electrocardiogram in
(PCAT)-2 Trialists’ Collaborative Group. Does time matter? A acute myocardial infarction. N Engl J Med 2003; 348:933–940.
[17/6/010/10:7:28] [429–443]
68. Gaitonde RS, Sharma N, Ali-Hasan S, et al. Prediction of signifi- 85. Ito H, Maruyama A, Iwakura K, et al. Clinical implications of the
cant left main coronary artery stenosis by the 12-lead electrocar- “no reflow” phenomenon: a predictor of complications and left
diogram in patients with rest angina pectoris and the withholding ventricular remodeling in reperfused anterior wall myocardial
of clopidogrel therapy. Am J Cardiol 2003; 92:846–848. infarction. Circulation 1996; 93:223–228.
69. Dixon WC IV, Wang TY, Dai D, et al. National Cardiovascular 86. Kelly RV, Cohen MG, Runge MS, et al. The no-reflow phenom-
Data Registry. Anatomic distribution of the culprit lesion in enon in coronary arteries. J Throm Haemost 2004; 2:1903–1907.
patients with non-ST-segment elevation myocardial infarction 87. Niccoli G, Burzotta F, Galiuto L, et al. Coronary no-reflow in
undergoing percutaneous coronary intervention: findings from humans. J Am Coll Cardiol 2009; 54:281–292.
the National Cardiovascular Data Registry. J Am Coll Cardiol 88. Sorajja P, Gersh BJ, Costantini C, et al. Combined prognostic
2008; 52:1347–1348. utility of ST-segment recovery and myocardial blush after pri-
70. Thompson CA. Transradial approach for percutaneous interven- mary percutaneous coronary intervention in acute myocardial
tion in acute myocardial infarction. J Invasive Cardiol 2009; 21 infarction Eur Heart J 2005; 26:667–674.
(suppl A):25A–27A. 89. Simes RJ, Topol EJ, Holmes DR Jr, et al. Link between the
71. Aviado DM, Aviado DG. The Bezold-Jarisch reflex. A historical angiographic substudy and mortality outcomes in a large
perspective of cardiopulmonary reflexes. Ann N Y Acad Sci 2001; randomized trial of myocardial reperfusion. Circulation 1995; 91:
940:48–58. 1923–1928.
72. Gibson CM, Cannon CP, Daley WL, et al. TIMI Frame Count. A 90. Galiuto L, Barchetta S, Paladini S, et al. Functional and structural
quantitative method of assessing coronary artery flow. Circula- correlates of persistent ST elevation after acute myocardial
tion 1996; 93:879–888. infarction successfully treated by percutaneous coronary inter-
73. Vlaar PJ, Svilaas T, van der Horst IC, et al. Cardiac death and vention. Heart 2007; 93:1376–1380.
reinfarction after 1 year in the thrombus aspiration during 91. Anderson JL, Karagounis LA, Califf RM. Meta-analysis of five
percutaneous coronary intervention in acute myocardial infarc- reported studies on the relation of early coronary patency grades
tion study (TAPAS): a 1-year follow-up study. Lancet 2008; 371: with mortality and outcomes after acute myocardial infarction.
1915–1920. Am J Cardiol 1996; 78:1–8.
74. Svilaas T, Vlaar PJ, van de Horst IC, et al. Thrombus aspiration 92. Goldman LE, Eisenberg MJ. Identification and management of
during percutaneous coronary intervention in acute myocardial patients with failed thrombolysis after acute myocardial infarc-
infarction. N Engl J Med 2008; 358:557–567. tion. Ann Intern Med 2000; 132:556–565.
75. De Luca G, Suryapranata H, Stone GW, et al. Adjunctive 93. Califf RM, O’Neill W, Stack RS, et al. Failure of simple clinical
mechanical devices to prevent distal embolization in patients measurements to predict perfusion status after intravenous
undergoing mechanical revascularization for acute myocardial thrombolysis. Ann Intern Med 1988; 108:658–662.
infarction: a meta-analysis of randomized trials. Am Heart J 2007; 94. Collet JP, Montalescot G, Le May M, et al. Percutaneous coronary
153:343–353. intervention after fibrinolysis: a multiple meta-analyses approach
76. Burzotta F, De Vita M, Gu YL, et al. Clinical impact of throm- according to the type of strategy. J Am Coll Cardiol 2006;
bectomy in acute ST-elevation myocardial infarction: an individ- 48:1326–1335.
ual patient-data pooled analysis of 11 trials. Eur Heart J 2009; 30: 95. Tcheng JE, Kinney KG. After fibrinolysis: Is cardiac catheter-
2193–2203. ization the answer? J Am Coll Cardiol 2006; 48:1336–1338.
77. Eid-Lidt G, Villavicencio-Fernandez R, Ponce De Leon-Rosales S, 96. Cantor WJ, Fitchett D, Borgundvaag B, et al. Routine early
et al. Comparative trial of a stent like balloon angioplasty versus angioplasty after fibrinolysis for acute myocardial infarction. N
coronary stenting for acute myocardial infarction. Catheter Car- Engl J Med 2009; 360:2705–2718.
diovasc Intervent 2001; 53:149–154. 97. Scheller B, Hennen B, Hammer B, et al. for the SIAM III Study
78. Mehta RH, Harjai KJ, Cox DA, et al. Primary Angioplasty in Group. Beneficial effects of immediate stenting after thrombol-
Acute Myocardial Infarction Investigators. Comparison of coro- ysis in acute myocardial infarction. J Am Coll Cardiol 2003; 42:
nary stenting versus conventional balloon angioplasty on five 634–641.
year mortality in patients with acute myocardial infarction 98. Le May MR, Wells GA, Labinaz M, et al. Combined angioplasty
undergoing primary percutaneous coronary intervention. Am J and pharmacological intervention versus thrombolysis alone in
Cardiol 2005; 96:901–906. acute myocardial infarction (CAPITAL AMI Study). J Am Coll
79. Grines CL, Cox DA, Stone GW, et al. Coronary angioplasty with Cardiol 2005; 46:417–424.
or without stent implantation for acute myocardial infarction. N 99. Fernandez-Avilés F, Alonso JJ, Castro-Beiras A, et al. on behalf of
Engl J Med 1999; 341:1949–1956. the GRACIA (Grupo de Análisis de la Cardiopatı́a Isquémica
80. Stone GW, Grines CL, Cox DA, et al. Comparison of angioplasty Aguda) Group. Routine invasive strategy within 24 hours of
with stenting, with or without abciximab, in acute myocardial thrombolysis versus ischemia-guided conservative approach
infarction. N Engl J Med 2002; 346:957–966. for acute myocardial infarction with ST-segment elevation
81. Van’t Hof AW, Liem A, de Boer MJ, et al. Clinical value of the 12- (GRACIA-1): a randomized controlled trial. Lancet 2004; 364:
lead electrocardiogram after successful reperfusion therapy for 1045–1053.
acute myocardial infarction. Zwolle Myocardial Infarction Study 100. Di Mario C, Dudek D, Piscione F, et al. Immediate angioplasty
Group. Lancet 1997; 117:18–24. versus standard therapy with rescue angioplasty after thrombol-
82. Brodie BR, Stuckey TD, Hansen C, et al. Intra-aortic balloon ysis in the combined abciximab reteplase stent study in acute
counterpulsation before primary percutaneous coronary angio- myocardial infarction (CARESS-in-AMI): an open, prospective,
plasty reduces catheterization laboratory events in high risk randomized, multicentre trial. Lancet 2008; 371:559–568.
patients with acute myocardial infarction. Am J Cardiol 1999; 84: 101. Thomas D, Giugliano RP. Management of ST-segment elevation
18–23. myocardial infarction: comparison of the updated guidelines
83. Stone GW, Marsalese D, Brodie BR, et al. A prospective random- from North America and Europe. Am Heart J 2009; 158:695–705.
ized evaluation of prophylactic intra-aortic balloon counterpul- 102. Van de Werf F, Bax J, Betriu A, et al. Management of acute
sation in high risk patients treated with primary angioplasty. myocardial infarction in patients presenting with persistent
Second Primary Angioplasty in Myocardial Infarction (PAMI-II) ST-segment elevation: the task force on the management of
Trial Investigators. J Am Coll Cardiol 1997; 29:1459–1467. ST-segment elevation acute myocardial infarction of the Euro-
84. Henriques JP, Zijlstra F, Ottervanger JP, et al. Incidence and clinical pean Society of Cardiology. Eur Heart J 2008; 29:2909–2945.
significance of distal embolization during primary angioplasty for
acute myocardial infarction. Eur Heart J 2002; 23:1112–1117.
[7/7/010/21:16:32] [444–452]
34
Cardiogenic shock
Mark J. Ricciardi
INTRODUCTION presence of vasopressor therapy, does not preclude the pres-

Circulatory shock is characterized by inadequate systemic tis- ence of CS. Together, these two frequently encountered clinical
sue perfusion due to altered physiology and reduced blood scenarios of preshock and mixed shock raise the possibility of
supply. Hypotension is differentiated from shock by the ability alternate mechanisms of hypotension in CS (such as systemic
to meet the tissue’s metabolic demands by maintaining tissue inflammation and vasodilatation) and iatrogenic contributions
perfusion. Tissue perfusion, a function of systemic vascular to the disorder (vasodilation with angiotensin-converting
resistance (SVR) and cardiac output, is significantly reduced in enzyme inhibitors, excessive diuresis and preload reduction,
instances where SVR is low and cardiac output is elevated and intravenous b-blocker therapy).
(vasodilatory shock) or where SVR is elevated and cardiac
output is reduced. When the latter is accompanied by inade-
quate intravascular volume, hypovolemic shock is present. ETIOLOGIES
Depressed cardiac output with elevated SVR in the presence The focus of the remainder of this chapter will be CS compli-
of adequate intravascular volume defines cardiogenic shock cating acute MI. Over 90% of cases fit into one of five etiologic
(CS) (Table 34.1). categories with approximately 80% of these due to pump
failure (left or right ventricular) and the remainder due to
myocardial rupture.
FUNDAMENTALS OF CS
The pathophysiology of CS due to pump failure usually begins
with profound depression of myocardial contractility. Resul- LV Pump Failure
tant reduction in cardiac output leads to hypotension, coronary LV pump failure is the dominant category of CS (Fig. 34.1) (6).
insufficiency, further reduction in contractility, and compensa- The vast majority of CS resulting from LV pump failure is due
tory systemic vasoconstriction. Without intervention, further to large, often anterior, infarction; a minority of cases is due to
declines in tissue perfusion, organ failure, and death may smaller infarctions in patients with preexisting LV dysfunction
ensue. Left ventricular (LV) pump failure is often due to resulting in cumulative loss of >40% of LV mass (4,7). Triple
large ST-segment elevation myocardial infarction (STEMI) but vessel or left main disease is present in 80% (8) and CS may be
can also be caused by smaller infarction in the presence of present despite only moderate degrees of LV systolic dysfunc-
preexisting LV dysfunction, right ventricular involvement, or tion (5). Data from the National Registry of Myocardial Infarc-
mechanical complications (papillary muscle, ventricular septal, tion (NRMI) database and others suggest that the incidence of
or free wall rupture). CS due to LV failure after MI is consistently 6% to 9% (9).
In clinical practice, two other scenarios are often encoun- However, data conflict regarding whether the incidence has
tered: preshock and mixed shock. Patients may develop a state decreased over the past few decades (10,11). The anecdotal
of preshock where hypoperfusion is present but compensatory impression of many clinicians that CS incidence is declining is
increases in heart rate and peripheral vasoconstriction allow for supported by observations published in 2008 analyzing data
the maintenance of cardiac output and systemic blood pressure. from 70% of the acute care hospitals in Switzerland. The
Preshock can be present at the time of hospital admission for observed downward trend was explained by decreases in CS
STEMI in patients who ultimately develop CS. Indeed, most not evident on presentation but developing during hospitaliza-
patients with CS complicating acute MI develop shock after tion (Fig. 34.2) (11).
initial presentation to the hospital (1). In the GUSTO-I (Global
Utilization of Streptokinase and Tissue Plasminogen Activator Clinical Manifestations
for Occluded Coronary Arteries) trial, five times the number of The typical clinical criteria and bedside findings of CS due to
patients developed shock after admission than they had it on LV pump failure are shown in Table 34.2. Physical examination,
presentation (2). In CS complicating non-STEMI, the delay may while helping rule out other causes for hypotension and shock,
be even more pronounced (3,4). The preshock state may pro- typically reveals ashen skin, cool extremities, altered senso-
vide a window of opportunity for aggressive reperfusion and rium, tachycardia, narrow pulse pressure, low pulse volume,
supportive strategies. In the case of mixed shock, evidence for distant heart sounds, and an audible and palpable left S3 gallop.
vasodilatation or underfilling confuses the classic paradigm of Distended neck veins may or may not be present; pulmonary
CS. In fact, the average SVR of patients in the SHOCK (Should congestion often is present. Urgent 12-lead electrocardiography
We Emergently Revascularize Occluded Coronaries for Cardio- usually shows STEMI or evidence for prior infarction with new
genic Shock) trial was not elevated with wide variability ST-segment depression, or left bundle branch block. Tachy- and
between patients (5). Normal SVR, therefore, even in the bradyarrhythmias are common.
[7/7/010/21:16:32] [444–452]
CARDIOGENIC SHOCK 445
Table 34.1 Shock Physiology

Mixed venous
Contractility Afterload Preload saturation Type of shock
: ; ; or $ : Vasodilatory
; : ; ; Hypovolemic
; : : ; Cardiogenic
Management
Addressing metabolic perturbations, respiratory insufficiency,
arrhythmias, and supportive care in general take on special
urgency with CS. If initial assessment suggests hypovolemia as
a possible etiology (i.e., no obvious pulmonary congestion), small
boluses of normal saline (100 cc at a time) may be used with
careful assessment of its effects on systemic blood pressure, heart
rate, and urine output. When CS (or preshock) is suspected,
quickly establishing the diagnosis and ruling out mechanical
complications with immediate echocardiographic, hemody-
namic, and angiographic evaluations is necessary. Because LV
failure CS is directly related to the degree of myocardial tissue
loss, strategies that limit infarct size are paramount. The benefits
of early revascularization for CS due to LV failure complicating
MI has been well described in the SHOCK trial (12). Comparing
early revascularization—either percutaneous coronary interven-
tion (PCI) or coronary artery bypass grafting (CABG)—with
initial medical stabilization with fibrinolysis, inotropes, and vaso-
pressor support when indicated, this landmark study demon-
strated numerical, albeit not statistically different, improvement
in 30-day survival with early revascularization (12) that became
statistically significant by six months and persisted for at least six
Figure 34.1 Etiologies of cardiogenic shock. Abbreviations: LVF,
predominant left ventricular failure; RVF, isolated right ventricular years (Fig. 34.3) (13). Intra-aortic balloon pump (IABP) counter-
shock; MR, acute severe mitral regurgitation; VSR, ventricular pulsation was used in 86% of patients. Those randomized within
septal rupture; FWR, free-wall rupture and cardiac tamponade. six hours of MI onset and those younger than 75 years had
Between groups, comparisons were based on hierarchical groups. particular benefit with early revascularization. Equally beneficial
Source: From Ref. 6. effects were seen with PCI (64%) and CABG (36%). While
unselected patients age >75 years in the SHOCK trial appeared
to have worse outcomes with early revascularization (13),
Figure 34.2 Temporal trends in the incidence of cardiogenic shock. Source: From Ref. 11.
[7/7/010/21:16:32] [444–452]
Table 34.2 Cardiogenic Shock due to LV Failure

Clinical criteria Bedside findings
Persistent systemic 30 minutes of systolic arterial
hypotension pressure <80 mmHg
Tissue hypoperfusion Oliguria, cold extremities, confusion
Reduced cardiac function Cardiac index <1.8 L/m2/min
Increased LV filling Pulmonary capillary wedge pressure
pressure >18 mmHg
Figure 34.5 IABP counterpulsation and thrombolytic therapy (TT)

for cardiogenic shock. Source: From Ref. 17.
Although relatively ineffective as a stand-alone therapy,

IABP counterpulsation is useful in stabilizing and temporizing
patients before more definitive therapy is instituted and during
the revascularization procedure. For patients with STEMI com-
plicated by CS who present to hospitals without primary
angioplasty capabilities, early fibrinolytic therapy and IABP
counterpulsation followed by immediate transfer for PCI or
Figure 34.3 Long-term survival after cardiogenic shock: Early
CABG is recommended (Fig. 34.5) (17). Critics may argue that a
revascularization versus initial medical stabilization. Source: From
Ref. 12. lack of randomized controlled mortality data supporting IABP
use exists. This lack of evidence is principally due to the early
acceptance into practice of this device for CS and the difficulty
of performing a meaningful trial testing its efficacy. A 2009
meta-analysis of over 10,000 patients with CS after acute MI
demonstrated some of the limitations of nonrandomized trials
(18). It showed that patients receiving fibrinolysis treated with
IABP had decreased 30-day mortality and those undergoing
primary PCI treated with IABP had increased mortality.
Despite attempts to account for confounders like younger age
and high PCI rates in the fibrinolytic/IABP group and attempts
to minimize selection bias, these data are not definitive. Many
would agree with the opinion that “as long as we do not have
any other evidence from randomized controlled clinical trials
we should not change our current practice because the benefi-
cial hemodynamic effects of the IABP should overweigh any
potential hazard” (19). In fact, the current guidelines from both
the European Society of Cardiology (ESC) and the American
College of Cardiology/American Heart Association (ACC/
AHA) give a class I recommendation for IABP in the setting
of CS (20,21). Despite this, only approximately one in four
patients with CS receive an IABP (9,22,23).
Although more complicated and costly, percutaneous
Figure 34.4 Survival for elderly patients undergoing PCI for car- cardiopulmonary bypass has also been used for CS. By provid-
diogenic shock. Source: From Ref. 15. ing up to 5 L/min of flow, percutaneous cardiopulmonary
bypass may improve survival but can be used for only several
hours because of hemolysis (24). Percutaneously placed extrac-
orporeal life support (ECLS) has also shown promise in select
nonrandomized data from the SHOCK Registry (14) and a sub- centers despite the gravest of clinical circumstances (25).
sequent study in 2009 (15) suggest that patients >75 years of age Although burdened by high complication rates, LV assist
who are otherwise good candidates for early revascularization devices provide better hemodynamic support than IABP but
may also benefit (Fig. 34.4) and should be considered for revas- have not been shown to improve survival in a small random-
cularization (16). ized trial (26). Presently, LV assist devices are best used as a
[7/7/010/21:16:32] [444–452]
Table 34.3 Indications for Pulmonary Artery Catheterization in Cardiogenic Shock

Should be performed (class I recommendation) if
l Progressive hypotension unresponsive to fluid or when fluid may be contraindicated
l Suspected mechanical complications of STEMI if echocardiogram not yet performed
Considered useful (class IIa recommendation) if
l Hypotension without pulmonary congestion not responsive to initial trial of fluid
l Cardiogenic shock
l Severe or progressive heart failure or pulmonary edema not responding to therapy
l Persistent signs of hypoperfusion without hypotension or pulmonary congestion
l During administration of vasopressor and/or inotropic agents
Abbreviation: ST-segment elevation myocardial infarction.
bridge to transplantation, and early transfer to centers provid- intravenously); the latter when a significant vasodilatory com-
ing such cardiopulmonary support and transplantation services ponent is present (5). Pure a-adrenergic agents such as phenyl-
is encouraged for appropriate candidates. In 2008, a much more ephrine should be reserved for the rare cases where
simply applied percutaneous LV assist device, the Impella 2.5, vasodilation predominates. The overzealous use of intravenous
was shown to offer superior hemodynamic support for CS b-blockers at the time of presentation for patients with acute
compared with IABP (27). The use of percutaneous assist coronary syndromes (ACS) and STEMI should be avoided. As
devices has not significantly displaced IABP use for CS outside noted, many CS patients initially present with preshock and
of a few select centers, and data suggesting more than just can easily deteriorate to CS by limiting both stroke volume and
hemodynamic supremacy over IABP are still lacking. While heart rate with acute b-blockade.
anecdotal experience is very encouraging, it is tempered by the
historical fact that improved hemodynamics do not necessarily Prognosis
translate into outcome advantages. A more comprehensive The in-hospital mortality rate for CS complicating MI remains
review of assist devices is covered elsewhere in this text. unacceptably high. During the 1970s and 1980s, mortality was
While controversy exists regarding the use of invasive >80% (30) and was only slightly better (70–75%) by the 1990s
hemodynamic monitoring in the critical care setting in general (31). The second NRMI analysis from hospitals with revascula-
(28,29), in expert hands the use of pulmonary artery catheter- rization capability documented decreasing CS mortality rates
ization for CS can be very helpful. The 2007 Focused Update of from 60% in 1995 to 48% in 2004 (9). The fall in mortality was
the 2004 ACC/AHA guidelines for the management of STEMI associated with an increase in revascularization rates, specifi-
states that pulmonary artery catheterization should be per- cally PCI, suggesting a possible causal relationship. The Swiss
formed, or is considered useful, in the majority of cases of CS MI registry published in 2008 had similar findings; overall CS
or suspected shock (Table 34.3) (20). in-hospital mortality decreased from 63% to 48% and rates of
Vasopressors and inotropes should be used to treat PCI and IABP increased, whereas rates of fibrinolytic therapy
hypotension. The agents to be considered include dopamine decreased and rates of CABG surgery remained stable
(5–15 mcg/kg/min intravenously), dobutamine (2–20 mcg/kg/ (Fig. 34.6) (11). Predictors of mortality from CS due to LV
min intravenously), and norepinephrine (0.5–30 mcg/min pump failure vary in the literature (6,32) probably due to
Figure 34.6 Trends in cardiogenic shock treatment. Source: From Ref. 11.
[7/7/010/21:16:32] [444–452]
differing patient populations studied, therapies used, and the while awaiting recovery of RV function after reperfusion. The
era during which the analyses were performed. Despite this, loss of atrioventricular synchrony can result in marked reduction
advanced age and concomitant multiorgan failure adversely in RV filling and systemic hypotension. Atrial or atrioventricular
affect survival, whereas the etiology of ACS does not (4). pacing or cardioversion may provide benefit in such circum-
stances (37). Typically, however, when patients are promptly
diagnosed, adequately supported, and treated with immediate
Right Ventricular Pump Failure reperfusion therapy, return of sinus rhythm and improved RV
Although some degree of right ventricular (RV) infarction is filling often is forthcoming and pacing unnecessary.
seen in up to 50% of inferior wall MI (33), resultant severe RV
dysfunction leading to CS is less common. Isolated RV infarc- Prognosis
tion and shock is rare (6). Somewhat surprisingly, predominant RV and LV shock
resulted in similar mortality in the SHOCK registry, despite
Clinical Manifestations younger age, less LV wall injury, and less multivessel disease in
The classic presentation of RV shock is that of CS without RV shock patients (34). Anecdotal experience and single center
pulmonary congestion, elevated right-sided filling pressures studies of patients undergoing primary PCI for predominant
(manifest by elevated jugular venous distention, Kussmaul’s RV shock, however, suggest they do better than their LV shock
sign, pulsus paradoxus, and pressure waveforms with steep counterparts (38), often recover quickly, and represent a par-
right atrial y decent and RV dip and plateau), ST-segment ticularly gratifying subgroup of shock patients to treat. In
elevation in the right precordial ECG leads, and RV chamber patients who do survive, return of RV function to normal is
enlargement and dysfunction. Marked depression in cardiac common (39).
output is present. On occasion, elevated right-sided filling
pressures can lead to right-to-left shunting through a patent
foramen ovale and manifest as unexpected degrees of hypo- Papillary Muscle Rupture
xemia. The clinical presentation of RV shock can include The three categories of myocardial rupture following MI have
components of LV dysfunction depending on the relative unique presentations (Table 34.4) and require mechanical inter-
involvement of the inferior LV. Patients with predominant vention for survival. Acute mitral regurgitation (MR) due to
RV shock are more likely to have single- or double-vessel papillary muscle rupture and dysfunction was responsible for
disease and less likely to have triple-vessel disease than 7% of CS in the SHOCK Registry of CS complicating acute MI (6).
patients with LV shock, and as expected, RV shock is rare
with acute left anterior descending occlusion (34). RV shock Clinical Manifestations
need not occur due to occlusion of the proximal right coronary The posteromedial papillary muscle is supplied by a single
artery (RCA) leading to cessation of blood flow to the RV source—the right or left circumflex coronary artery. Inferior MI
marginal artery. It can be seen in conjunction with posterior due to thrombosis of either of these arteries can therefore lead
LV infarction and distal RCA or distal dominant circumflex to posteromedial papillary necrosis and rupture. Because the
occlusion (35,36). dual sourced anterolateral papillary muscle is relatively pro-
tected from necrosis after thrombotic coronary occlusion, ante-
Management rior wall MI less commonly leads to acute MR. Acute MR shock
Immediate reperfusion therapy, either with fibrinolysis or PCI, is typically presents with profound hypotension and pulmonary
the goal of therapy. Avoidance of medications that decrease edema. The lack of an audible murmur is not uncommon and
preload (nitroglycerin or morphine) or depress cardiac output does not rule out acute MR due to papillary rupture. Immediate
(b-blockers) is mandatory. Maintaining or augmenting RV pre- echocardiography for suspected acute MR is diagnostic. Dislo-
load with volume expansion constitutes initial therapy for pre- cation of the tip of the papillary muscle or choral rupture is
dominant RV shock. If hypotension is not reversed after 1 to 2 L often seen. As expected, LV ejection fraction may be better
of fluid, cardiac output augmentation with dobutamine may be preserved than those with CS due to LV failure. The pulmonary
necessary, especially if there is a delay to reperfusion therapy or capillary wedge tracing may show tall V waves.
Table 34.4 Cardiogenic Shock due to Myocardial Rupture
PMR VSR FWR

Location 3/4 inferior 2/3 anterior 1/2 anterior
Exam Murmur in 50% Murmur in 90% JVD, PEA
Thrill Rare Common No
Echocardiography Regurgitant jet Shunt Effusion
PA catheterization C-V wave in wedge tracing O2 step-up Diastolic equalization
Abbreviations: JVD, jugular venous distension; PEA, pulsus electrical activity; PMR, VSR, ventricular septal rupture; FWR, free-wall rupture.
[7/7/010/21:16:32] [444–452]
Ventricular Septal Rupture

Ventricular septal rupture (VSR) complicating MI was respon-
sible for 3% of CS in the SHOCK Registry of CS complicating
acute MI (6) and represents a particularly challenging group
because of the difficulties of operative management and very
poor prognosis.
Clinical Manifestations
Patients with VSR complicating MI often fit the classic rupture
dictum “first MI in an older hypertensive patient presenting
late” (42,43). In the SHOCK registry, VSR patients were less
likely to have had prior MI compared to those with LV failure
CS, and the average age was 72 years (43). Unlike acute MR,
VSR is almost always accompanied by a murmur (harsh
holosystolic) and thrill. Doppler echocardiography visualizes
Figure 34.7 In-hospital mortality for cardiogenic shock due to the shunt, and oxygen saturation step up in the RV is present.
mitral regurgitation. Abbreviation: MV, mitral valve. Source: From Both anterior and inferior infarction can lead to VSR.
Ref. 41.
Management
CS due to VSR is a nearly uniformly fatal ailment requiring
Management immediate surgery. Delaying surgery in patients who do not
Immediate mitral valve surgery (repair or replacement) with have overt CS is appealing in concept as it may allow for
revascularization is the only currently established treatment. healing of the defect margins. Unfortunately, waiting risks the
All other efforts should be aimed at stabilizing the patient prior development of shock, which is unpredictable and significantly
to surgery. IABP counterpulsation should be used and, if the worsens surgical survival. Because of the high mortality of
augmented systemic blood pressure allows, afterload reduction immediate surgical repair, some have experimented with less
with intravenous sodium nitroprusside should be started. invasive acute percutaneous VSR closure using septal occlusion
Patients with papillary muscle necrosis and rupture (and not devices. While still in its infancy, this less invasive approach to
simply ischemia) would not be expected to benefit from revas- treating acute VSR does not appear to offer an advantage over
cularization alone, and primary PCI does not appear to benefit surgery, especially when CS is present (44).
the degree of MR in the setting of severe acute papillary
rupture complicating STEMI (40). Prognosis
VSR with CS carries with it one of the worst survival rates of all
Prognosis cardiac conditions (Fig. 34.8) (6). Despite high operative mor-
Like all cases of myocardial rupture complicating MI, survival tality, surgical repair offers an improved chance of survival
with or without aggressive surgical intervention is poor. In the (Fig. 34.9) (43). Although not all surgical series agree (45), VSR
nonrandomized SHOCK registry, patients undergoing mitral complicating inferior MI can be particularly challenging for two
valve surgery for acute MR fared poorly, albeit significantly reasons. First, because RV involvement is more likely with
better than those managed medically (Fig. 34.7) (41). inferior infarction, the hemodynamics are more foreboding.
Figure 34.8 Mortality rates for the major shock categories. Abbreviations: LVF, predominant LV failure; RVF, isolated RV shock; MR, acute
severe mitral regurgitation; VSR, ventricular septal rupture; Tamp, free-wall rupture and cardiac tamponade. Source: From Ref. 6.
[7/7/010/21:16:32] [444–452]
bed rest, and blood pressure control with b-blocker therapy) for
patients surviving initial FWR and tamponade (51); however,
prompt surgical repair is recommended.
Prognosis
Many patients succumb almost instantly with rapid and irre-
versible electromechanical dissociation. Patients presenting
with subacute FWR and hemopericardium appear to have
survival rates similar to CS due to LV pump failure (48,49).
Survival requires prompt recognition of tamponade, aggressive
stabilization, and prompt surgery.
CONCLUSIONS
The high lethality of CS in conjunction with the importance of
establishing the correct etiology makes the management of CS
Figure 34.9 In-hospital mortality for cardiogenic shock due to challenging and fascinating. While the two types of pump
ventricular septal rupture. Source: From Ref. 43. failure CS are best treated with urgent (usually percutaneous)
revascularization, stabilizing medical therapies differ signifi-
cantly and require early detection of LV versus RV failure.
Conversely, the three types of myocardial rupture CS share
Second, proximal RCA occlusion leading to basal septal VSR is similar supportive therapies but usually require very special-
more difficult to repair than the apical defects often seen with ized and complex surgical interventions; urgent percutaneous
anterior MI (46). In making decisions about whether to proceed revascularization is rarely indicated and may squander valu-
to surgery in patients with many comorbidities and advanced able time best used to establish a diagnosis and prepare for
age, the location of the VSD may help determine the usefulness surgery.
or futility of going forward with surgery.
REFERENCES
Free-Wall Rupture and Hemorrhagic Tamponade 1. Webb JG, Sleeper LA, Buller CE, et al. Implications of the timing of
The most devastating complication of MI, free-wall rupture onset of cardiogenic shock after acute myocardial infarction: a
(FWR), was responsible for 1% of CS in the SHOCK Registry of report from the SHOCK Trial Registry. SHould we emergently
CS complicating acute MI (6). revascularize Occluded Coronaries for cardiogenic shocK? J Am
Coll Cardiol 2000; 36:1084–1090.
Clinical Manifestations 2. Holmes DR Jr., Bates ER, Kleiman NS, et al. Contemporary
reperfusion therapy for cardiogenic shock: the GUSTO-I trial
Like other forms of myocardial rupture complicating MI, FWR
experience. The GUSTO-I Investigators. Global Utilization of
often presents with a first MI and may be more common in Streptokinase and Tissue Plasminogen Activator for Occluded
hypertensive women (42,47). Typically, the infarct is large and Coronary Arteries. J Am Coll Cardiol 1995; 26:668–674.
anterior (47,48). Because acute FWR and hemopericardium can 3. Hasdai D, Harrington RA, Hochman JS, et al. Platelet glycoprotein
lead to abrupt hemodynamic collapse, electromechanical dis- IIb/IIIa blockade and outcome of cardiogenic shock complicating
sociation, and near instant death, many of the clinical manifes- acute coronary syndromes without persistent ST-segment eleva-
tations of FWR reported in the literature may represent a tion. J Am Coll Cardiol 2000; 36:685–692.
subacute variety of FWR in which the rupture is contained. A 4. Holmes DR Jr., Berger PB, Hochman JS, et al. Cardiogenic shock
spectrum exists from catastrophic to subacute, with clinical in patients with acute ischemic syndromes with and without
presentation varying from immediate death to hypotension ST-segment elevation. Circulation 1999; 100:2067–2073.
5. Hochman JS. Cardiogenic shock complicating acute myocardial
with pericarditic chest pain. Whether fibrinolytic therapy
infarction: expanding the paradigm. Circulation 2003; 107:
increases the likelihood of rupture is in dispute and was not 2998–3002.
found to be the case in the SHOCK trial registry (48,49). The 6. Hochman JS, Buller CE, Sleeper LA, et al. Cardiogenic shock
lack of pulmonary edema is more likely in FWR than LV pump complicating acute myocardial infarction—etiologies, manage-
failure CS (17% vs. 55%) and may provide a diagnostic clue ment and outcome: a report from the SHOCK Trial Registry.
(49). Echocardiography has obvious utility in diagnosis and is SHould we emergently revascularize Occluded Coronaries for
very specific for FWR when the pericardial effusion is large cardiogenic shocK? J Am Coll Cardiol 2000; 36:1063–1070.
(50). Because small effusions and tamponade physiology may 7. Holmes DR Jr. Cardiogenic shock: a lethal complication of acute
be missed, up to 25% of patients with FWR may have no myocardial infarction. Rev Cardiovasc Med 2003; 4:131–135.
significant pericardial effusion by echocardiography (49,50). 8. Sanborn TA, Sleeper LA, Webb JG, et al. Correlates of one-year
survival inpatients with cardiogenic shock complicating acute
myocardial infarction: angiographic findings from the SHOCK
Management
trial. J Am Coll Cardiol 2003; 42:1373–1379.
Early recognition and swift operative repair are the keys to
9. Babaev A, Frederick PD, Pasta DJ, et al. Trends in management
management. Contained ruptures may allow for the institution and outcomes of patients with acute myocardial infarction com-
of supportive measures like IABP counterpulsation and vaso- plicated by cardiogenic shock. JAMA 2005; 294:448–454.
pressor therapy in preparation for surgery. Some patients can 10. Goldberg RJ, Samad NA, Yarzebski J, et al. Temporal trends in
be temporized with pericardiocentesis. Long-term survival cardiogenic shock complicating acute myocardial infarction.
without surgery has been reported (using pericardiocentesis, N Engl J Med 1999; 340:1162–1168.
[7/7/010/21:16:32] [444–452]
11. Jeger RV, Radovanovic D, Hunziker PR, et al. Ten-year trends in 27. Seyfarth M, Sibbing D, Bauer I, et al. A randomized clinical trial to
the incidence and treatment of cardiogenic shock. Ann Intern Med evaluate the safety and efficacy of a percutaneous left ventricular
2008; 149:618–626. assist device versus intra-aortic balloon pumping for treatment of
12. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization cardiogenic shock caused by myocardial infarction. J Am Coll
and long-term survival in cardiogenic shock complicating acute Cardiol 2008; 52:1584–1588.
myocardial infarction. JAMA 2006; 295:2511–2515. 28. Connors AF Jr., Speroff T, Dawson NV, et al. The effectiveness of
13. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization in right heart catheterization in the initial care of critically ill patients.
acute myocardial infarction complicated by cardiogenic shock. SUPPORT Investigators. JAMA 1996; 276:889–897.
SHOCK Investigators. Should We Emergently Revascularize 29. Harvey S, Young D, Brampton W, et al. Pulmonary artery cathe-
Occluded Coronaries for Cardiogenic Shock. N Engl J Med 1999; ters for adult patients in intensive care. Cochrane Database Syst
341:625–634. Rev 2006; 3:CD003408.
14. Dzavik V, Sleeper LA, Cocke TP, et al. Early revascularization is 30. Goldberg RJ, Gore JM, Alpert JS, et al. Cardiogenic shock
associated with improved survival in elderly patients with acute after acute myocardial infarction. Incidence and mortality from a
myocardial infarction complicated by cardiogenic shock: a report community-wide perspective, 1975 to 1988. N Engl J Med 1991; 325:
from the SHOCK Trial Registry. Eur Heart J 2003; 24:828–837. 1117–1122.
15. Lim HS, Farouque O, Andrianopoulos N, et al. Survival of elderly 31. Goldberg RJ, Gore JM, Thompson CA, et al. Recent magnitude of
patients undergoing percutaneous coronary intervention for acute and temporal trends (1994–1997) in the incidence and hospital
myocardial infarction complicated by cardiogenic shock. J Am death rates of cardiogenic shock complicating acute myocardial
Coll Cardiol Intervent 2009; 2:146–152. infarction: the second national registry of myocardial infarction.
16. Hochman JS, Skolnick AH. Contemporary management of cardio- Am Heart J 2001; 141:65–72.
genic shock: age is opportunity. J Am Coll Cardiol Intervent 2009; 32. Singh M, White J, Hasdai D, et al. Long-term outcome and its
2:153–155. predictors among patients with ST-segment elevation myocardial
17. Sanborn TA, Sleeper LA, Bates ER, et al. Impact of thrombolysis, infarction complicated by shock: insights from the GUSTO-I trial.
intra-aortic balloon pump counterpulsation, and their combina- J Am Coll Cardiol 2007; 50:1752–1758.
tion in cardiogenic shock complicating acute myocardial infarc- 33. Kinch JW, Ryan TJ. Right ventricular infarction. N Engl J Med
tion: a report from the SHOCK Trial Registry. SHould we 1994; 330:1211–1217.
emergently revascularize Occluded Coronaries for cardiogenic 34. Jacobs AK, Leopold JA, Bates E, et al. Cardiogenic shock caused
shocK? J Am Coll Cardiol 2000; 36:1123–1129. by right ventricular infarction: a report from the SHOCK registry.
18. Sjauw KD, Engstrom AE, Vis MM, et al. A systematic review and J Am Coll Cardiol 2003; 41:1273–1279.
meta-analysis of intra-aortic balloon pump therapy in ST-eleva- 35. Andersen HR, Falk E, Nielsen D. Right ventricular infarction:
tion myocardial infarction: should we change the guidelines? Eur frequency, size and topography in coronary heart disease: a pro-
Heart J 2009; 30:459–468. spective study comprising 107 consecutive autopsies from a cor-
19. Thiele H, Schuler G. Cardiogenic shock: to pump or not to pump? onary care unit. J Am Coll Cardiol 1987; 10:1223–1232.
Eur Heart J 2009; 30:389–390. 36. Isner JM, Roberts WC. Right ventricular infarction complicating
20. Antman EM, Hand M, Armstrong PW, et al. 2007 Focused Update of left ventricular infarction secondary to coronary heart disease.
the ACC/AHA 2004 Guidelines for the Management of Patients Frequency, location, associated findings and significance from
With ST-Elevation Myocardial Infarction: a report of the American analysis of 236 necropsy patients with acute or healed myocardial
College of Cardiology/American Heart Association Task Force on infarction. Am J Cardiol 1978; 42:885–894.
Practice Guidelines: developed in collaboration With the Canadian 37. Topol EJ, Goldschlager N, Ports TA, et al. Hemodynamic benefit
Cardiovascular Society endorsed by the American Academy of of atrial pacing in right ventricular myocardial infarction. Ann
Family Physicians: 2007 Writing Group to Review New Evidence Intern Med 1982; 96:594–597.
and Update the ACC/AHA 2004 Guidelines for the Management of 38. Brodie BR, Stuckey TD, Hansen C, et al. Comparison of late
Patients With ST-Elevation Myocardial Infarction, Writing on Behalf survival in patients with cardiogenic shock due to right ventric-
of the 2004 Writing Committee. Circulation 2008; 117:296–329. ular infarction versus left ventricular pump failure following
21. Van de Werf F, Bax J, Betriu A, et al. Management of acute primary percutaneous coronary intervention for ST-elevation
myocardial infarction in patients presenting with persistent acute myocardial infarction. Am J Cardiol 2007; 99:431–435.
ST-segment elevation: the Task Force on the Management of 39. Dell’Italia LJ, Lembo NJ, Starling MR, et al. Hemodynamically
ST-Segment Elevation Acute Myocardial Infarction of the Euro- important right ventricular infarction: follow-up evaluation of
pean Society of Cardiology. Eur Heart J 2008; 29:2909–2945. right ventricular systolic function at rest and during exercise
22. Barron HV, Every NR, Parsons LS, et al. The use of intra-aortic with radionuclide ventriculography and respiratory gas exchange.
balloon counterpulsation in patients with cardiogenic shock Circulation 1987; 75:996–1003.
complicating acute myocardial infarction: data from the National 40. Tcheng JE, Jackman JD Jr., Nelson CL, et al. Outcome of patients
Registry of Myocardial Infarction 2. Am Heart J 2001; 141:933–939. sustaining acute ischemic mitral regurgitation during myocardial
23. Zeymer U, Vogt A, Zahn R, et al. Predictors of in-hospital mor- infarction. Ann Intern Med 1992; 117:18–24.
tality in 1333 patients with acute myocardial infarction compli- 41. Thompson CR, Buller CE, Sleeper LA, et al. Cardiogenic shock
cated by cardiogenic shock treated with primary percutaneous due to acute severe mitral regurgitation complicating acute myo-
coronary intervention (PCI); Results of the primary PCI registry of cardial infarction: a report from the SHOCK Trial Registry.
the Arbeitsgemeinschaft Leitende Kardiologische Krankenhau- SHould we use emergently revascularize Occluded Coronaries
sarzte (ALKK). Eur Heart J 2004; 25:322–328. in cardiogenic shocK? J Am Coll Cardiol 2000; 36:1104–1109.
24. Nichol G, Karmy-Jones R, Salerno C, et al. Systematic review of 42. Figueras J, Cortadellas J, Calvo F, et al. Relevance of delayed
percutaneous cardiopulmonary bypass for cardiac arrest or car- hospital admission on development of cardiac rupture during
diogenic shock states. Resuscitation 2006; 70:381–394. acute myocardial infarction: study in 225 patients with free wall,
25. Bartlett RH, Roloff DW, Custer JR, et al. Extracorporeal life support: septal or papillary muscle rupture. J Am Coll Cardiol 1998; 32:
the University of Michigan experience. JAMA 2000; 283:904–908. 135–139.
26. Thiele H, Sick P, Boudriot E, et al. Randomized comparison of 43. Menon V, Webb JG, Hillis LD, et al. Outcome and profile of
intra-aortic balloon support with a percutaneous left ventricular ventricular septal rupture with cardiogenic shock after myocardial
assist device in patients with revascularized acute myocardial infarction: a report from the SHOCK Trial Registry. SHould we
infarction complicated by cardiogenic shock. Eur Heart J 2005; emergently revascularize Occluded Coronaries in cardiogenic
26:1276–1283. shocK? J Am Coll Cardiol 2000; 36:1110–1116.
[7/7/010/21:16:32] [444–452]
44. Thiele H, Kaulfersch C, Daehnert I, et al. Immediate primary 49. Slater J, Brown RJ, Antonelli TA, et al. Cardiogenic shock due to
transcatheter closure of postinfarction ventricular septal defects. cardiac free-wall rupture or tamponade after acute myocardial
Eur Heart J 2009; 30:81–88. infarction: a report from the SHOCK Trial Registry. Should we
45. Deja MA, Szostek J, Widenka K, et al. Post infarction ventricular emergently revascularize occluded coronaries for cardiogenic
septal defect—can we do better? Eur J Cardiothorac Surg 2000; shock? J Am Coll Cardiol 2000; 36:1117–1122.
18:194–201. 50. Lopez-Sendon J, Gonzalez A, Lopez de Sa E, et al. Diagnosis of
46. Moore CA, Nygaard TW, Kaiser DL, et al. Postinfarction ventric- subacute ventricular wall rupture after acute myocardial infarc-
ular septal rupture: the importance of location of infarction and tion: sensitivity and specificity of clinical, hemodynamic and
right ventricular function in determining survival. Circulation echocardiographic criteria. J Am Coll Cardiol 1992; 19:1145–1153.
1986; 74:45–55. 51. Figueras J, Cortadellas J, Evangelista A, et al. Medical management
47. Shapira I, Isakov A, Burke M, et al. Cardiac rupture in patients of selected patients with left ventricular free wall rupture during
with acute myocardial infarction. Chest 1987; 92:219–223. acute myocardial infarction. J Am Coll Cardiol 1997; 29:512–518.
48. Honan MB, Harrell FE Jr., Reimer KA, et al. Cardiac rupture,
mortality and the timing of thrombolytic therapy: a meta-analysis.
J Am Coll Cardiol 1990; 16:359–367.
[2/7/010/9:18:30] [453–468]
35
A clinical rationale and strategy for chronic total coronary

occlusion revascularization: innovative solutions
to an old problem
Colin M. Barker and David E. Kandzari
INTRODUCTION (TIMI grade 1 flow), referred to as a “functional” total occlu-

Despite remarkable advances in the procedural and clinical sion. Without serial angiograms, the duration of coronary
outcomes of percutaneous coronary intervention (PCI), chron- occlusion is difficult to specify with any certainty and must
ically occluded coronary arteries remain a formidable challenge be estimated from available clinical information related to the
and unresolved dilemma in interventional cardiology. timing of the event that caused the occlusion, for example,
Although a total coronary occlusion is identified in approxi- clinical history of myocardial infarction (MI) or sudden change
mately one-third of diagnostic cardiac catheterizations, an in angina pattern with electrocardiographic changes consistent
attempted revascularization accounts for less than 8% of all with the location of the occlusion. In most patients, the age of
PCIs (1). In the Emory Angioplasty versus Surgery Trial the CTO cannot be determined with precision. Furthermore, the
(EAST), for example, the presence of a chronic total occlusion temporal criterion used to define a CTO has varied among
(CTO) was the most common reason for physician referral to registries, trials, and databases, ranging from >2 weeks (6–8) to
coronary artery bypass graft surgery (CABG) (2). Such a dis- >3 months (9), which in part explains interstudy differences in
parity between their frequency and treatment not only under- lesion characteristics and procedural success. In general, a total
scores the technical and procedural frustrations associated with occlusion of duration >3 months may be considered “chronic.”
these complex lesions, but also the clinical uncertainties regard-
ing which patients may benefit from CTO revascularization.
CTOs remain the single most important reason not to attempt CTO Histopathology
PCI in favor of CABG or medical treatment. Illustrating this, in CTOs most often arise from thrombotic occlusion followed by
the multivessel PCI versus CABG SYNTAX (Synergy between thrombus organization and tissue aging (10,11). Particularly
PCI with TAXUS and Cardiac Surgery) study, the prevalence of relevant to PCI strategies for CTO recanalization is the histo-
CTO in the randomized arm was 27% compared with 59% in logical finding that approximately half of all CTOs are <99%
the CABG registry cohort (3). stenotic when observed by histopathology, despite the angio-
Unlike PCI of nonocclusive coronary disease, much of graphic appearance of total occlusion with TIMI grade 0 ante-
our understanding regarding CTOs has been further chal- grade flow. Moreover, little to no relationship exists between
lenged by relatively few studies describing the procedural the severity of the histopathological lumen stenosis and either
and clinical outcomes among patients undergoing attempted plaque composition or lesion age.
revascularization. Moreover, these investigations are limited by The typical atherosclerotic plaque of a CTO consists of
their retrospective, observational design, variability in operator intracellular and extracellular lipids, smooth muscle cells,
skills, and inconsistencies regarding the definition of CTO and extracellular matrix, and calcium. Collagens are the major
patient selection. Since the duration of an occluded artery is an structural components of the extracellular matrix (8,12), with
independent predictor of procedural outcome (4,5), an inability predominance of types I and III (and minor amounts of IV, V,
to date these lesions, in addition to their heterogeneous com- and VI) in the fibrous stroma of atherosclerotic plaques (13).
position, has restricted the evaluation of novel revasculariza- The concentration of collagen-rich fibrous tissue is particularly
tion technologies. Until recently, many of the technologies dense at the proximal and distal ends of the lesion, contributing
promoted for the treatment of CTOs were simply modeled to a column-like lesion of calcified, resistant fibrous tissue
after devices applied to nonocclusive disease, erroneously surrounding a softer core of organized thrombus and lipids.
assuming that the pathophysiology between these lesion sub- Key histopathological attributes of CTOs are calcification
sets were similar. extent, inflammation, and neovascularization. The typical CTO
may be classified as “soft,” “hard,” or a mixture of both (Figs. 35.1
and 35.2). Soft plaque consists of cholesterol-laden cells and
ANANTOMIC CONSIDERATIONS foam cells with loose fibrous tissue and neovascular channels,
The definition of a coronary CTO is reflective of the degree of and is more frequent in younger occlusions (<1-year-old). Soft
lumen stenosis, the amount of antegrade blood flow, and the plaque is more likely to allow guidewire passage either directly
age of the occlusion. The degree of stenosis can range from through tissue planes or via neovascular channels into the
complete occlusion [Thrombolysis in Myocardial Infarction distal lumen. Conversely, hard plaques are characterized by
(TIMI) grade 0 flow], frequently called a “true” CTO, to 99% dense fibrous tissue and often contain large fibrocalcific regions
occlusion with minimal contrast penetration through the lesion without neovascular channels. During PCI, these occlusions are
[2/7/010/9:18:30] [453–468]
thus more likely to deflect coronary guidewires into the sub-

intimal area, creating dissection planes. Hard plaques are more
prevalent with increasing CTO age (>1-year-old). Of note,
however, areas of calcification frequently occur even in CTOs
<3 months of age, although the extent and severity of calcifi-
cation increase with occlusion duration. This age-related
increase in calcium and collagen content of CTOs in part
underlies the progressive difficulty during PCI in crossing
older occlusions.
Inflammatory cell infiltrates in CTOs consist of macro-
phages, foam cells, and lymphocytes. Inflammation may exist
in the intima, media, and adventitia of CTOs, although it is
most predominant in the intima. As fibrotic CTO lesions age,
the vessels typically undergo negative remodeling with
decreasing dimension of the external elastic membrane, a phe-
nomenon due to adventitial vascular responses. Occasionally,
plaque hemorrhage and inflammation may result in positive
remodeling (14).
Another hallmark of CTOs is extensive neovasculariza-
tion, which occurs throughout the extent of the vessel wall.
Capillary density and angiogenesis increase with advancing
Figure 35.1 (See color insert) (A) CTO, soft plaque (hematoxylin- occlusion age. In CTOs <1-year-old, new capillary formation is
eosin stain; magnification 1). (B) Magnified view of A, showing greatest in the adventitia. In CTOs >1-year-old, the number
cholesterol clefts and loose fibrous tissue (hematoxylin-eosin stain; and size of capillaries in the intima have increased to a similar
magnification 10). (C) CTO, hard plaque, dense fibrous tissue, and or greater extent than those present in the adventitia. Relatively
calcium (elastic van Gieson stain; magnification 1). Abbreviation: large (>250 mm) capillaries are frequently (47–67%) present
CTO, chronic total occlusion. Source: From Ref. 8. throughout the CTO vessel wall, even in young occlusions,
suggesting that angiogenesis within the CTO is an early event.
Frequent colocalization of inflammation and neovasculariza-
tion within the intimal plaque and adventitia suggests that
Figure 35.2 (See color insert) (A) A single large channel is seen in this CTO (elastic van Gieson stain; magnification 1). (B) Traversing
capillaries connect with the small recanalization channels in the center of this CTO (elastic van Gieson stain; magnification 1). (C) Small
recanalization vascular channels are seen in the center of this CTO (elastic van Gieson stain; magnification 1). (D) Inflammation is found
adjacent to vascular channels of the adventitia in this vessel (hematoxylin-eosin stain; magnification 25). (E) Adventitial capillaries have
grown to large size in this CTO (hematoxylin-eosin stain; magnification 40). Abbreviation: CTO, chronic total occlusion. Source: From Ref. 8.
[2/7/010/9:18:30] [453–468]
A CLINICAL RATIONALE AND STRATEGY FOR CTO REVASCULARIZATION 455
these findings are closely related, although it is unclear whether Recovery of impaired left ventricular function after revas-
inflammation is a cause or an effect of neovascularization in cularization of a CTO is not directly related to the quality of
CTOs. Lymphocytes and monocytes/macrophages may play collateral function, as collateral development does not appear
an active role in both angiogenesis and atherosclerotic lesion to require the presence of viable myocardium. However, those
progression by producing a variety of mitogenic and angio- patients with recovery had a lower peripheral resistance mea-
genic factors (15). sured by intracoronary Doppler and pressure wires as an
A rich neovasculature network often traverses the CTO indicator of preserved microvascular integrity (19).
vessel wall, arising from the adventitial vasa vasorum across
the media and into the lesion intima, suggesting that vessel Clinical Impact Relative to Target Vessel
ingrowth proceeds from the adventitia in younger lesions. An Very little data exist regarding the potential for differential
autopsy study of subtotal atherosclerotic lesions demonstrated benefit of CTO recanalization depending on the target vessel,
that new intimal vessels originate in the adventitial vasa vaso- for example, left anterior descending (LAD), left circumflex
rum of lesions with >70% stenosis, but rarely from the coronary (LCX), or right coronary artery (RCA). A recent, large, single-
lumen (16). Such microchannels, which can recanalize the distal center registry suggests that PCI for CTO of the LAD, but not
lumen, may result from thrombus-derived angiogenic stimuli LCX or RCA, is associated with improved long-term survival
(17) and are suggested on an angiogram of an old CTO without (20). This study included 2608 patients and the LAD was the
a well-defined proximal cap or stump. In this regard, the target vessel in 936 (36%), the LCX in 682 (26%), and the RCA in
distinction should be made between ipsilateral epicardial 990 (38%) patients. Angiographic success rates were 77%, 76%,
angiographic “bridging” collateral vessels and true microvas- and 72%, respectively. Procedural success compared with fail-
cular collaterals. Neochannels may also develop with organi- ure was associated with improved five-year survival in the
zation of thrombus, connecting the proximal and distal lumens; LAD cohort (88.9% vs. 80.2%, P < 0.001), but not in the LCX
this is suggested by a tapered CTO proximal cap on an angio- (86.1% vs. 82.1%, P ¼ 0.21) and RCA groups (87.7% vs. 84.9%,
gram. Such channels may serve as a route for a guidewire to P ¼ 0.23). In multivariable analysis, CTO PCI success in the
reach the distal vessel and hence may have therapeutic value. LAD group remained associated with decreased mortality risk
[HR 0.61; 95% confidence interval (CI), 0.42–0.89]. In addition to
other clinical characteristics, this information may assist in
Collaterals and CTOs
selecting patients for attempted CTO PCI.
Collaterals preserve myocardial function and avoid cell death
in the distribution of the occluded artery. A CTO that is well
collateralized is functionally equivalent to a 90% stenosis (18) INDICATIONS FOR REVASCULARIZATION
(Fig. 35.3). The myocardium remains viable but produces In general, when the CTO represents the only significant lesion
ischemia during periods of increased oxygen demand, and in the coronary tree, PCI is warranted when the following three
thus patients with these lesions are likely to have exertional conditions are all present (8):
angina. The risk of unstable angina or an acute coronary syn- 1. The occluded artery is responsible for the patient’s symp-
drome due to the lesion is unlikely as it is totally occluded. The toms of chest pain or heart failure despite maximal medical
myocardium supplied by the CTO can result in an acute cor- therapy (PCI may also be considered in selected cases of
onary syndrome if the arteries supplying the collaterals become silent ischemia if a large myocardial territory at risk is
compromised in any way. demonstrable).
2. The CTO territory is associated with significant left ventric-
ular dysfunction and viability is demonstrated within the
myocardial territory supplanted by the occluded artery.
3. The likelihood of success is moderate to high (>60%), with
an anticipated major complication rate of death <1% and
MI <5%.
If the PCI attempt is unsuccessful, further management
will depend on the symptomatic status and the extent of
jeopardized ischemic myocardium. Repeated PCI following
initial failure (typically with an allowance of several weeks
for vessel healing in the case of dissection) or CABG may be
warranted if a large myocardial territory is ischemic or the
patient is very symptomatic. Alternatively, conservative ther-
apy may be appropriate if repeated PCI is unlikely to be
successful and the patient’s symptoms can be managed with
antianginal medications.
In patients with multivessel disease and one or more
CTOs, the relative risks and benefits of CABG compared with
PCI should be considered. The presence of any of the following
may favor CABG (8):
Figure 35.3 Angiogram with the injection of contrast into the left 1. Left main artery disease
coronary artery showing the extensive collaterals supplying the right 2. Complex triple-vessel disease, especially in the patient
coronary artery. with insulin-requiring diabetes, severe left ventricular
dysfunction, or chronic kidney disease
[2/7/010/9:18:30] [453–468]
3. An occluded proximal LAD (supplying a viable anterior reported no improvement in left ventricular function at one
wall), which is not favorable for PCI year with revascularization. Given these findings, it may be
4. Multiple CTOs with a relatively low anticipated success best to allow recovery from acute MI for patients similar to
rate. those included in this trial who present to the interventional
suite within this time period. Conversely, a more recent meta-
Other patients with multivessel disease, including a CTO,
analysis of randomized control trials demonstrated that PCI of
may be appropriately managed by PCI, with the goal being
the infarct artery performed late (12 hours to 60 days) after MI
complete revascularization whenever possible. Typically, in
is associated with significant improvements in cardiac function
patients presenting with an acute coronary syndrome or with
and survival (29). Angiographic documentation of total occlu-
stable angina in whom the nonoccluded vessels can be reliably
sion was required as an inclusion criterion in 5 of the 10 studies,
stented with a low rate of complication, angioplasty of the CTO
and 3 additional studies required either a total occlusion or
should be performed after PCI of nonoccluded lesions. The one
significant stenosis. The analysis included 3560 patients with
exception to this rule would be in patients in whom failed PCI
median time from MI to randomization of 12 days (range 1–26
of the CTO would result in referral for CABG, in which case the
days) and follow-up of 2.8 years (42 days to 10 years). Ran-
CTO should be approached first unless conditions dictate
domization allocated 1779 subjects to PCI and 1781 to medical
otherwise.
treatment. There were 112 (6.3%) PCI and 149 (8.4%) medical
Patients with medically refractory angina or a moderate
therapy deaths, yielding significantly improved survival in the
to large ischemic burden deserve consideration for PCI, partic-
PCI group (OR 0.49; 95% CI, 0.26–0.94; P ¼ 0.030). These
ularly if the symptoms or jeopardized myocardial territory are
benefits were associated with similarly favorable effects on
enough to warrant CABG as an option. In several large data-
cardiac remodeling, including improved left ventricular ejec-
bases, only 11% to 15% of patients undergoing PCI for CTO
tion fraction in the PCI group (þ4.4% change; 95% CI, 1.1–7.6;
were asymptomatic (21,22). Conversely, the proportion of
P ¼ 0.009).
patients presenting with unstable angina due to a CTO is also
fairly low (9–18%) (8,23). Thus, the majority of patients under-
going PCI for CTO have stable or progressive angina, whereas PROCEDURAL FUNDAMENTALS
many asymptomatic patients with CTO are managed medi- The technical and procedural success rates of PCI in CTOs have
cally. A history of prior MI has been reported in 42% to 68% of modestly increased over the last 20 years because of greater
patients with an angiographically demonstrated CTO. operator experience and improvements in equipment and pro-
Stress-induced ischemia can typically be elicited in cedural techniques (30). Despite this observation, CTOs remain
patients with CTOs, especially in the absence of a history of the lesion subtype in which PCI is most likely to fail. In recent
prior MI. One study identified reversible perfusion defects contemporary series, procedural success rates have ranged
using stress myocardial single-photon emission computed from 55% to 80%, with the variability reflecting differences in
tomography (SPECT) in 83% of 71 patients without history of operator technique and experience, availability of advanced
prior MI and with single-vessel disease involving a CTO (24). guidewires, CTO definition, and case selection. The most com-
Similarly, another study documented severe and extensive mon PCI failure mode for CTOs is inability to successfully pass
stress perfusion defects in 56 patients with no prior MI and a guidewire across the lesion into the true lumen of the distal
single-vessel CTO with the presence of collaterals (25). Adeno- vessel.
sine SPECT imaging may be even more sensitive than exercise- Most studies have consistently reported that increasing
induced stress imaging to detect perfusion defects in patients age of the occlusion, greater lesion length, presence of a non-
with CTO. tapered stump, origin of a side branch at the occlusion site,
The temporal changes in contractility and hyperemic and excessive vessel and lesion tortuosity, calcification, ostial occlu-
resting myocardial blood flow (MBF) in dependent and remote sion, and lack of visibility of the distal vessel course negatively
myocardium after PCI of CTOs were investigated using car- affect the ability to successfully cross a CTO (30). In the past,
diovascular magnetic resonance (CMR) imaging (26). Three the presence of bridging collaterals was also consistently iden-
groups were prospectively studied: 17 patients scheduled for tified as a determinant of failed CTO PCI, but in more recent
CTO PCI, 17 scheduled for PCI of a nonocclusive coronary experiences, success rates may no longer be inversely corre-
artery stenosis (non-CTO), and 6 patients with CTO who were lated with the presence of bridging collaterals and other angio-
not scheduled for revascularization. Contractility in treated graphic characteristics. Bridging collaterals may reflect the
segments was improved at 24 hours and 6 months after CTO chronicity of the lesion and signify the requirement for stiffer
PCI but only at 6 months after non-CTO PCI. In both PCI and/or tapered tip wires to penetrate the occlusion. The avail-
groups, treated segments no longer had reduced MBF or con- ability of enhanced force wires with greater torque response
tractility compared with remote segments. In patients with has clearly increased the success rates in occlusions with
untreated CTO segments, however, MBF and wall thickening bridging collaterals and similar angiographic complexities to
did not improve at follow-up. the point where CTOs should no longer be avoided for these
Debate and conflicting data exist regarding the optimal reasons alone.
timing for revascularization of a total coronary occlusion result- A recent multicenter CTO registry identified predictors of
ing from a recent infarction. In the OAT (Occluded Artery Trial) unsuccessful recanalization (31). This study documented expe-
(27), PCI of infarct arteries 3 to 28 days after MI did not reduce riences with 1362 patients at three centers from 2000 to 2007.
death, recurrent MI, or congestive heart failure over 4 years. Both angiographic and clinical outcomes were measured over
Notably, exclusion criteria included multivessel coronary dis- long-term follow-up, presently out to three years for the major-
ease, rest angina, or demonstration of at least moderate ische- ity of patients. Recanalization was successful in 65.5% of native
mia by noninvasive stress testing. The TOSCA-2 (Total vessels and 76.6% of in-stent restenosis cases. In the overall
Occlusion Study of Canada) substudy (28) of this trial also cohort, a longer lesion, blunt appearance of the proximal
[2/7/010/9:18:30] [453–468]
occlusion, vessel calcification, and prior CABG were all asso- the procedure, decrease complications, and ultimately improve
ciated with procedural failure. procedural outcome (32).
To increase the likelihood of success, an 8-Fr guiding
catheter is recommended. Extended length sheaths (35 cm)
CLINICAL AND PROCEDURAL CONSIDERATIONS will provide additional support. Larger guide catheters provide
FOR CTO REVASCULARIZATION excellent support for guidewire penetration but may also be
Imaging: Invasive Angiography, CT Angiography, associated with increased contrast utilization. Larger diameter
and IVUS guiding catheters also enable delivery of covered stent grafts in
Once a decision has been made to attempt recanalization of a uncommon instances of coronary perforation after successful
CTO, characterization of the plaque (e.g., extent of calcification, recanalization. If a second angiographic catheter is used for
CTO origin) and visualization of the distal vessel must be contralateral injection, a 5- or 6-Fr catheter may be inserted into
optimized. In most instances, this may be achieved with the contralateral femoral artery or either brachial or radial
invasive angiography utilizing contralateral injection of con- artery. For RCA CTOs, an Amplatz guide will provide maximal
trast into the artery supplying collaterals to the distal vessel support, while any supportive catheter with extra backup can
(Fig. 35.4). If there is any doubt regarding the location of the be used for the left coronary system. One exception is in
true lumen or the anatomical course of the occluded vessel instances using the retrograde approach (described below), in
segment, a coronary CT angiogram with 3D reconstruction may which a short guide (85 cm) is required in the “donor” artery
be particularly useful (Fig. 35.5). Furthermore, in selected cases, that supplies the collateral flow.
coronary CT angiography may improve patient selection for Intravascular ultrasound (IVUS) has become a routine
CTO recanalization, decrease the time and contrast needed for imaging modality in interventional cardiology and can be very
useful in the evaluation and treatment of CTOs. Using an
antegrade approach to the lesion, IVUS can be used to
1. identify the proximal cap location using the IVUS catheter
in the side branch,
2. confirm the wire penetration into the proximal cap,
3. redirect the wire into the true lumen after penetrating the
subintimal space, and
4. optimize stent placement, expansion, and apposition.
When performing a retrograde approach, IVUS can be
used to help guide the retrograde wire into the proximal lumen
of the CTO (Fig. 35.6).
Antithrombotic Therapy
Unfractionated heparin (UFH) is the preferred antithrombin
therapy during percutaneous CTO revascularization. In the
case of coronary perforation, unlike presently available direct
thrombin inhibitors, UFH can be readily reversed with admin-
istration of protamine (10 mg reverses 1000 U of UFH). In most
instances, a reduced initial bolus dose is administered to achieve
an activated clotting time (ACT) of approximately 200 to
Figure 35.4 Angiogram with dual injections showing a chronic total 250 seconds until the guidewire has successfully crossed the
occlusion (CTO) of the left anterior descending artery (LAD). CTO, after which additional UFH may be required before angio-
plasty and stent placement to achieve an ACT of approximately
Figure 35.5 CT angiogram of a known chronic total

occlusion (CTO) of the right coronary artery (RCA)
demonstrating heavy calcification at the site of occlu-
sion and a significant bend in the RCA within the
occluded segment.
[2/7/010/9:18:30] [453–468]
be steered more easily in a true lumen immediately after sharp

bends. However, they are more likely than noncoated wires to
penetrate beneath plaque and cause subintimal dissections.
Furthermore, they less commonly maintain their tip shape
compared with nonpolymer coated wires and do not offer
precise tip control. A hydrophilic wire may be passed for
long distances in a false channel without resistance. The course
of hydrophilic wires must be assessed in at least two orthog-
onal views to confirm the wire is in the true lumen. The most
commonly used polymer-coated wires are the Asahi Fielder
(Regular, FC, and XT), the Abbott Whisper, the Guidant Pilot
(ranging in support from 50 to 200), and the Boston Scientific
Choice and Graphix series. Once a CTO-specific guidewire has
crossed the occlusion and passed into the distal lumen, the wire
should be exchanged with a soft, floppy-tipped wire to mini-
mize the risk of distal wire perforation or dissection.
A microcatheter or an over-the-wire (OTW) 1.5-mm bal-
loon catheter can be used for support as well as access for ease
of exchanging wires. The current 1.5-mm compliant balloons
Figure 35.6 Intravascular ultrasound of the ostial left main coro- are very low profile and are able to cross lesions in a majority of
nary artery showing a retrograde guidewire penetrating from the cases. Smaller-diameter (1.25 mm) balloons are presently
subintimal space into the aorta (white arrow). under investigation and are available outside the United States.
A balloon catheter also offers the option of treatment with
dilation of the vessel as well as added support by using it as
an anchoring device within the vessel for greater guidewire
support.
300 seconds [in the absence of intravenous glycoprotein (GP) Inability to cross a CTO with a balloon catheter despite
IIb/IIIa inhibitor use]. Bivalirudin and GP IIb/IIIa inhibitors are successful guidewire recanalization is an infrequent event.
usually avoided at the start of the procedure because of the Assuming adequate guiding catheter support, one method is
increased procedural risk of bleeding complications (e.g., tam- to use a second angioplasty balloon in a side branch vessel
ponade) in instances of coronary perforation. If indicated, proximal to the CTO (“anchoring balloon” technique). In some
GP IIb/IIIa inhibitors may be administered once the guidewire instances, incremental inflations with a small diameter
has successfully crossed the lesion and is confirmed to be intra- (1.5 mm) balloon within the proximal segment of a CTO
luminal. Pretreatment with clopidogrel has anecdotally not been may facilitate eventual crossing. Other treatment options
associated with increased risk of CTO procedural complications include laser atherectomy using a 0.9-mm laser catheter, rota-
but has not been well studied. Similar to other PCI procedures, if tional atherectomy (which requires passage of specialized
the patient is naive to thienopyridine therapy, a loading dose rotablator guidewire), or the Tornus (Asahi Intec, Aichi,
may be administered during or at the end of the procedure. Japan) support catheter. In the latter example, the Tornus
catheter is a braided stainless steel support catheter that is
advanced in a counterclockwise fashion over the guidewire to
Guidewire Technology and Selection create a passage in highly stenosed lesions and facilitate sub-
Crossing the CTO with a guidewire is the most important and sequent balloon catheter delivery.
challenging step of the procedure and is the most frequent
cause for failed PCI of a CTO. There are three separate steps to
crossing a CTO: penetrating the proximal fibrous cap; travers- Guidewire Techniques: Antegrade, Retrograde,
ing the body of the CTO to reach the distal fibrous cap; and Controlled Dissection
penetrating the distal fibrous cap to enter the true lumen. Wires The antegrade approach is the most common initial strategy for
designed particularly for treating CTOs can be broadly divided attempting to recanalize a CTO. Starting equipment may vary
into two major groups: polymer-jacketed and/or hydrophilic depending on lesion characteristics and anatomy. A support
guidewires, and stiffer, nonpolymer (nonhydrophilic or hydro- catheter (OTW balloon or microcatheter) with either a stiff
philic) guidewires. Both wire categories may be available in nonhydrophilic guidewire (e.g., Miracle Bros 3 g, Asahi
standard (0.014-in. diameter) or tapered (0.009- to 0.010-in. Intec/Abbott Vascular, Redwood City, California, U.S.) or a
diameter) tip configurations. The stiffer, nonhydrophilic guide- tapered hydrophilic wire (e.g., Fielder XT, Abbott Vascular) is
wires are typically more controllable, provide better tactile feel, used to probe the lesion initially. If the proximal cap cannot be
and are less likely to cause vessel dissection. Techniques differ, penetrated or the wire is unable to be advanced within the
but depending on operator experience, it is generally recom- lesion, progressively stiffer wires may be substituted. For some
mended to initially attempt CTO penetration with a less stiff experienced CTO interventionalists, high penetrating force
guidewire and progress to graduated tip-load wires if resis- wires (e.g., 9- to 12-g tip load) are often selected as an initial
tance to penetration is encountered. guidewire. With the use of the stiffer (and hydrophilic) guide-
Hydrophilic wires advance with minimal resistance and wires, however, the operator must be aware of the increased
tactile feel, even down minute branches and false channels, risk of complications (e.g., coronary perforation or major dis-
contributing to an increased risk of coronary perforation. These section). Wire shaping for the antegrade approach of CTOs is
guidewires offer maneuverability in tortuous vessels and may markedly different than for nonocclusive disease. In general,
[2/7/010/9:18:30] [453–468]
the initial wire should have a bend with an angle <308 approx-
imately 1 mm from the tip. If more angulated bends are
required to access the CTO, this should be done with a soft,
floppy wire and then exchanged for a dedicated CTO wire.
Two specialized techniques for CTO recanalization using
an antegrade approach have evolved for failure after initial
wire entry in the lesion. The parallel wire technique is a com-
mon method to approach CTO PCI (Fig. 35.7). Occasionally a
guidewire may exit the true lumen of the occluded vessel and
enter a subintimal dissection plane. In this instance, the wire is
left in place as a visual landmark to avoid further vessel trauma
and to obstruct entry into the false lumen. A second wire (stiff
or hydrophilic, depending on the lesion characteristics) is
advanced to the point of the first wire’s exit and then redirected
toward the true lumen. A twin-pass dual access exchange
catheter (Vascular Solutions, Inc., Minneapolis, Minnesota,
U.S.) can be very useful when introducing the second wire.
An alternative method is the subintimal tracking and reentry
(STAR) technique (33) (Fig. 35.8). This is a method of inten-
Figure 35.7 Angiogram of a CTO of the right coronary artery using tional subintimal dissection using a hydrophilic wire typically
a parallel wire technique. The first guidewire entered the subintimal shaped with a prolapsed tip. The wire is advanced beyond the
space and was left in place. A second guidewire was passed occlusion adjacent to the distal lumen and may spontaneously
adjacent to the first one and directed toward the true lumen. reenter the true lumen in the main vessel, or more commonly at
Abbreviation: CTO, chronic total occlusion.
the confluence of bifurcating side branches. Ultimately, an
extensive dissection is created, which typically requires
Figure 35.8 Angiogram showing the subintimal tracking and reentry (STAR) technique for treatment of a right coronary artery CTO (A). This
is a method of intentional subintimal dissection using a hydrophilic wire with a prolapsed tip. The wire is advanced beyond the occlusion
adjacent to the distal lumen (B). It may spontaneously reenter the true lumen in the main vessel, or more commonly in side branches.
Ultimately, an extensive dissection is created, which needs to be treated with several stents (C,D). Abbreviation: CTO, chronic total occlusion.
[2/7/010/9:18:30] [453–468]
Figure 35.9 Angiogram showing a retrograde approach for treatment of a CTO of the left anterior descending artery (A). After selecting a
septal collateral, a polymer-coated wire is advanced to the distal edge of the occlusion (B). The proximal lumen is accessed and treated with
angioplasty. Next, guidewires are introduced in from an antegrade approach (C), and the vessel is treated definitively (D). Abbreviation: CTO,
chronic total occlusion.
treatment with several stents. Notably, this technique may not is essential to define the location, size, and tortuosity of the
only be associated with a higher risk of coronary perforation vessels. Once localized, the collateral can be crossed with a
but can also result in shearing and occlusion of side branches supportive hydrophilic wire such as a Fielder FC (Asahi) and
and should be reserved as a bailout technique for highly microcatheter. To deliver support catheters or angioplasty
symptomatic patients refractory to medical therapy. balloons to the target vessel, the septal collateral often requires
More complex techniques for difficult CTOs include the dilatation. This can be accomplished with either a dedicated
retrograde approach as well as the controlled antegrade and septal dilator catheter (Corsair, Asahi), or a 1.25- to 1.5-mm
retrograde subintimal tracking (CART) and reverse CART diameter compliant balloon at very low atmospheres. A stiff
methods (34,35) (Fig. 35.9). These advanced techniques are wire, such as a Miracle Bros 3 g or Confianza, is then used to
generally reserved for passing wires in a retrograde course penetrate the proximal cap via a retrograde approach followed
through septal or occasionally epicardial collaterals, although by balloon dilatation and then antegrade crossing. Alterna-
manipulations in epicardial collaterals are extremely high risk tively, in cases of successful retrograde crossing but inability
for perforation. Because the collaterals often originate from the to deliver balloon catheters, the retrograde wire (300 cm in
contralateral coronary artery, dual arterial catheter access is length) may be externalized through the antegrade guiding
required, and shortened guiding catheters (85–90 cm) are essen- catheter and then exchanged and followed by antegrade PCI.
tial (at least for the retrograde donor catheter) to enable appro- IVUS from the proximal portion of the CTO may also assist in
priate working length. Selective angiography of the collaterals guiding the wire into the true lumen.
[2/7/010/9:18:30] [453–468]
Table 35.1 Randomized Clinical Trials of Angioplasty Versus Stenting for Chronic Total Coronary Occlusions
Reocclusion Restenosis TVR
PTCA Stent PTCA Stent PTCA Stent
Trial N (%) (%) P (%) (%) P (%) (%) P
Stenting in Chronic Coronary 114 26 16 0.058 74 32 <0.001 42 22 0.025
Occlusion (SICCO) (36)
Gruppo Italiano di Studi sulla 110 34 8 0.004 68 32 0.0008 22 5 0.04
Stent nelle Occlusioni
coronariche (GISSOC) (37)
Mori et al. (38) 96 11 7 0.04 57 28 0.005 49 28 <0.05
Stent vs Percutaneous 85 24 3 0.01 64 32 0.01 40 25 NS
Angioplasty in Chronic
Total Occlusion (SPACTO) (39)
Total Occlusion 410 20 11 0.02 70 55 <0.01 15 8 0.03
Study of Canada (TOSCA) (40)
Stents in Total Occlusion for 96 17 8 NS 71 42 0.032 42 25 NS
Restenosis Prevention
(STOP) (41)
Abbreviations: NS, not significant; PTCA, percutaneous transluminal coronary angioplasty.
In cases of primary retrograde crossing failure, variations BMS. Additional studies have demonstrated statistically signif-
on the retrograde approach include controlled dissection tech- icant yet modest improvement in left ventricular function and
niques termed CART and reverse CART. With CART, a balloon regional wall motion with CTO revascularization (37,46–48).
is inflated in the distal portion of the CTO over the retrograde Importantly, an increase in left ventricular function may be
wire that is located in the false lumen. A second antegrade wire conditional on vessel patency at follow-up and revasculariza-
is introduced distal to the proximal cap and into the subintimal tion of total occlusions of shorter duration (6 weeks) (49).
space. When the retrograde balloon is inflated, expanding the Except for lesion length (49), exactly which features pre-
subintimal space, the antegrade wire is then directed toward dict late target lesion failure remain elusive (50). While the
the retrograde balloon as it is deflated. Once the antegrade wire lesion complexity of CTOs cannot be altered, the potential for
enters the space occupied by the balloon, it may then be new stent designs to improve rates of restenosis and reocclu-
advanced distally into the true lumen parallel to the path sion is considerable. Two observations mandate the need for
taken by the retrograde wire and balloon. The reverse CART systematic evaluation of drug-eluting stents (DES) in CTO
is the same concept, although the roles are reversed; specifi- revascularization: (i) advances in CTO technical and procedural
cally, a balloon is inflated in the proximal portion of the CTO success have been disproportionate to the increasing number of
on the antegrade wire, and the retrograde wire is advanced into PCI procedures that involve nonacute occlusions and (ii) out-
the subintimal space. These techniques are highly specialized comes with stenting in CTOs are more similar to balloon
and carry a significant risk. They require dedicated equipment angioplasty than to stenting in nonocclusive lesions.
and unique skills learned from extensive training, separate
from most other interventional procedures.
Drug-Eluting Stents
Sirolimus-Eluting Stents
CORONARY STENT OUTCOMES IN CTO Against the background of several nonrandomized, observa-
REVASCULARIZATION tional studies demonstrating improved angiographic and clin-
Six randomized trials of bare-metal stent (BMS) placement ical outcomes with DES, only one randomized trial comparing
compared with balloon angioplasty alone have been reported DES with BMS has been performed. In the Primary Stenting of
(Table 35.1). Collectively, these trials have shown that BMS Occluded Native Coronary Arteries (PRISON) II trial, 200 CTO
implantation achieves statistically significant and clinically patients were randomized in a single-blinded fashion at two
meaningful reductions in angiographic restenosis and reocclu- centers in The Netherlands to treatment with either sirolimus-
sion. Compared with angioplasty alone, BMS also confer a eluting stents (SES; Cypher, Cordis Corporation, Bridgewater,
long-term benefit with fewer repeat revascularization proce- New Jersey, U.S.) or the bare-metal BX Velocity stent (Cordis
dures (23,42). Despite these results, outcomes with BMS in Corporation) (51). Patients enrolled in this study underwent
CTOs are more similar to outcomes with balloon angioplasty six-month angiographic follow-up to assess the primary end-
in less complex, nonocclusive disease (43,44). In the TOSCA point of in-segment binary restenosis (50% reduction in min-
trial (40), for example, rates of restenosis and reocclusion imal lumen diameter). Overall, diabetes mellitus was present in
in complex lesions exceeded 50% and 10%, respectively. At 13% of patients, 55% of patients had a total occlusion <3 months
three-year follow-up, the occurrence of reocclusion was asso- old, and the average lesion and stent lengths were approxi-
ciated with a trend toward higher mortality and significant mately 16 mm and 30 mm, respectively. At six months, treat-
increase in the need for repeat revascularization (45). ment with SES was associated with statistically significant
Although these trials have varied considerably regarding reductions in both in-stent (7% vs. 36%; P < 0.0001) and in-
enrollment criteria, antithrombotic regimen, and trial design, segment (11% vs. 41%; P < 0.001) angiographic restenosis
their results are remarkably concordant, demonstrating signif- (Fig. 35.10). Reocclusion was also significantly reduced with
icant reductions in restenosis and reocclusion associated with SES (4% vs. 13%; P < 0.04), despite treatment in both groups
[2/7/010/9:18:30] [453–468]
In the Rapamycin-Eluting Stent Evaluated at Rotterdam

Cardiology Hospital (RESEARCH) registry, among 56 patients
treated with SES following CTO revascularization, the one-year
occurrence of TVR was 3.6%, compared with 17.9% among a
historical control group of patients receiving BMS (54). Simi-
larly, the six-month rate of TLR was only 1.4% for 360 patients
with CTOs who were included in the prospective e-Cypher
registry (53). Among 180 patients undergoing SES implantation
for CTO revascularization in Asia, the six-month occurrences of
angiographic binary restenosis and TVR were 1.5% and 2.3%,
respectively (62).
As part of a multicenter Asian registry evaluating DES,
clinical and angiographic outcomes among 60 patients who
underwent SES implantation during CTO revascularization
were compared with a matched control of 120 CTO patients
treated with BMS (62). At six-month angiographic follow-up,
Figure 35.10 Six-month angiographic results from the PRISON II treatment with SES was associated with significant reductions
Trial. ABR denotes angiographic binary restenosis. Abbreviations: in in-stent late loss, restenosis, and reocclusion. TLR was
TLR, target lesion revascularization; MACE, major adverse cardiac significantly lower at six months (23.0% vs. 2.0%; P ¼ 0.001),
events; SES, sirolimus-eluting stents. Source: Adapted from Ref. 51. and the left ventricular ejection fraction also significantly
improved among the SES patients (51.8% baseline vs. 57.0%
at 6 months; P < 0.01). This latter finding implies that mainte-
nance of vessel patency with DES may be an important pre-
with aspirin and clopidogrel for a minimum duration of six dictor of the improvement in left ventricular function. At one
months. The clinical benefit with SES also paralleled the relative year, treatment with SES was associated with sustained reduc-
benefit observed with angiographic measures. Specifically, target tions in hierarchal MACE (3.0% vs. 42.0%; P ¼ 0.001) and TLR
lesion revascularization (TLR) at 12 months occurred in 21% and (43.0% vs. 3.0%; P ¼ 0.001). In a related study of 226 patients
5% of BMS and SES patients, respectively (P ¼ 0.001). undergoing CTO revascularization (SES 106, BMS 120), treat-
In addition to this randomized trial, several recent modest- ment with SES was associated with a sustained, significant
sized observational studies examining clinical outcomes among reduction in MACE through four-year follow-up (7.5% vs.
patients treated with DES in CTO revascularization have sup- 33.8%; P < 0.001) (56).
ported the notion that DES may achieve similar reductions in the Most recently, the Approaches to Chronic Occlusions
need for target vessel revascularization (TVR), as observed in with Sirolimus-Eluting Stents/Total Occlusion Study of Coro-
less complex lesions (Table 35.2). In a retrospective study of nary Arteries-4 (ACROSS)/TOSCA-4 trial prospectively
122 patients with CTOs treated with SES (N ¼ 144 lesions), enrolled 200 patients undergoing CTO revascularization with
clinical and angiographic outcomes were compared with a his- SES using contemporary techniques and crossing technologies
torical control of 259 patients treated with BMS (N ¼ 286 lesions) (Fig. 35.11) (61). In this nonrandomized study, clinical and
(54). At six months, overall major adverse cardiac events six-month angiographic outcomes were compared with a his-
(MACE) were significantly lower with SES (16.4% vs. 35.1%; torical control of patients receiving BMS in the prior TOSCA-1
P < 0.001), principally due to a significantly lower rate of TLR trial. Compared with the BMS group, patients treated with SES
(7.4% vs. 26.3%; P < 0.001). Restenosis was identified in 9.2% of had significantly older age, more CTOs (i.e., >6 weeks), smaller
patients in the SES group and 33.3% in the BMS group (P < 0.001). caliber vessels, a higher proportion of diabetes, and longer
In multivariate analysis, significant predictors of six-month MACE lesion and stent lengths. However, despite higher complexity
were the use of BMS (HR 2.97; 95% CI, 1.80–4.89), lesion length in the SES cohort, treatment with SES was associated with an
(HR 2.02; 95% CI, 1.37–2.99), and reference vessel diameter unadjusted 66% relative reduction in the primary endpoint of
>2.8 mm (HR 0.62; 95% CI, 0.42–0.92). angiographic binary restenosis within the treated segment
Table 35.2 Clinical Trials Evaluating Drug-Eluting Stents in Total Coronary Occlusions
6 mo 1 yr
Angiographic
Trial N restenosis (%) TVR (%) MACE (%) TVR (%) MACE (%)
SICTO (52) 25 0 8.0 12.0 12.0 12.0
e-Cypher Registry (53) 360 – 1.4a 3.1 – –
RESEARCH Registry (54) 56 9.1 3.6 3.6 – –
Werner et al. (55) 48 8.3 – – 6.3 12.5
Nakamura et al. (56) 60 2.0 3.0 – 3.0
Ge et al. (57) 122 9.2 9.0 16.4 – –
WISDOM Registry (58) 65 – – – 6.7 1.7
TRUE Registry (59) 183 17.0 16.9 17.1 – –
Buellesfeld et al. (60) 45 13.2 13.2 15.6 – –
ACROSS/TOSCA-4 (61) 200 7.5 6.0 6.5 – –
a
Denotes target lesion revascularization.
Abbreviations: TVR, target vessel revascularization; MACE, major adverse cardiac events.
[2/7/010/9:18:30] [453–468]
Figure 35.11 Rates of six-month angiographic in-stent binary restenosis for selected patient subgroups. Abbreviation: DM, diabetes
mellitus. Source: Adapted from Ref. 61.
(19% vs. 55%; P < 0.001), defined as the length of contiguous with CTOs in the international WISDOM registry, treatment
target segment exposed to balloon dilation prior to stent place- with PES resulted in freedom from MACE and TVR at one
ment. Following adjustment for baseline characteristics predic- year in 93.3% and 98.3% of patients, respectively (58).
tive of restenosis, the treatment effect increased to an 84% More recently, analyses related to the subgroup of CTO
relative reduction in treated segment restenosis. Rates of inpatients included in the SYNTAX trial have been reported (52).
segment and in-stent restenosis were 11.5% and 6.5%, respec- Conducted at 62 European sites and 23 sites in the United
tively. At six-month clinical follow-up, rates of MI and TLR States, the SYNTAX trial randomized 1800 patients to either
were 1% and 6%, respectively, contributing to a 6% occurrence CABG (N ¼ 897) or PCI (N ¼ 903) with PES to examine a
of the composite endpoint of target vessel failure (cardiovas- primary endpoint of 12-month MACE and cerebrovascular
cular death, MI, or TVR). events (MACCE), defined as all-cause death, cerebrovascular
event, MI, and any repeat revascularization (PCI and/or
CABG). Altogether, 479 patients had CTOs (CABG 235, PCI
Paclitaxel-Eluting Stents
244) that contributed to a higher SYNTAX lesion complexity
Compared with studies evaluating SES, relatively limited evi-
score. Overall, less than 50% of the CTOs were successfully
dence exists to support the use of paclitaxel-eluting stents (PES)
treated with PCI, but if any revascularization attempt was
in CTO revascularization. One study examined treatment with
made, the success rate was 78%. Interestingly, despite random-
PES (Taxus, Boston Scientific Corporation, Natick, Massachu-
ization to CABG, more than 30% of vessels involving a CTO
setts, U.S.) in 48 patients undergoing CTO PCI and compared
were not surgically revascularized. Safety outcomes (all death,
these patients with a historical control group with similar
stroke, MI) for patients with CTOs were similar between PCI- or
clinical and angiographic characteristics (55). At six months,
CABG-treated patients, although CTO lesions treated with PCI
both restenosis (8.3% vs. 51.1%; P < 0.001) and reocclusion
had a higher rate of revascularization compared with CABG.
(2.1% vs. 23.4%; P < 0.005) were significantly reduced among
patients treated with PES. Because of significantly decreased
rates of repeat PCI or CABG, MACE was also significantly Comparative DES Trials in CTO Revascularization
lower in the PES group. One year following the index revascu- Whether safety, clinical efficacy, and angiographic outcomes
larization, repeat revascularization occurred in 3 patients in the are similar between differing DES has only been recently
PES group and 21 patients in the BMS group (6.3% vs. 43.8%; examined (63,64). Despite more predictable variance in mea-
P < 0.001). sures of neointimal hyperplasia by angiography and IVUS,
In the European TRUE Registry among 183 patients with demonstration of differences in clinical outcome across indi-
CTOs who were treated with PES, seven-month rates of reste- vidual trials has been less consistent (65–74). However, whether
nosis and TVR were 17.0% and 16.9%, respectively (59). It is disparities in angiographic and clinical outcome emerge in
noteworthy that the mean (standard deviation) number of more complex lesion morphologies is an issue of ongoing
stents per patient (2.2 1.2) and total stent length (58 33 mm) study and is particularly relevant to CTOs.
were considerably greater than previous DES trials involving At present, four comparative trials of SES and PES have
treatment of less complex lesion subsets. Other studies evalu- been performed (Table 35.3). In general, these studies have
ating PES in CTO PCI have included fewer patients. One study been limited by their small study populations that limit statis-
examined clinical and angiographic outcomes among 45 CTO tical comparisons, variability in trial design, and limited clinical
patients treated with PES (60). At six months, the rates of and angiographic follow-up. In the single-center Rotterdam
angiographic restenosis and TVR were 13.2%. Among 65 patients registry (RESEARCH and T-SEARCH) comparing clinical
[2/7/010/9:18:30] [453–468]
Table 35.3 Comparative Drug-Eluting Stent Trials in Total Coronary Occlusions

Angiographic restenosis (%) TVR (%) MACE (%)
Trial N (SES:PES) SES PES SES PES SES PES
RESEARCH/T-SEARCH 76:57 – – 2.6 3.6 – –
Registry (75)a
Nakamura et al.
Asian Registry (76)a 396:526 4.0 6.7 3.6 6.7 3.6 6.7
de Lezo et al. (77)b,c 60:58 7.4 19.0 3.3 7.0 3.0 7.0
Jang et al. (78) 107:29 9.4 28.6d 3.7 6.9 4.2 14.2d
P, not significant for all comparisons unless otherwise noted.
a
One-year outcomes.
b
Eight-month outcomes.
c
Angiographic follow-up in only 48% of patients.
d
P < 0.05.
Abbreviations: TVR, target vessel revascularization; MACE, major adverse cardiac events; SES, sirolimus-eluting stents; PES, paclitaxel-
eluting stents.
outcomes among CTO patients treated with BMS (N ¼ 26), SES not occur at the occluded segment or be related to the guide-
(N ¼ 76) and PES (N ¼ 57), one-year freedom from repeat TVR wire itself rather than to adjunctive devices including balloon
was significantly greater with DES compared with BMS (97.4% inflation, stent implantation, or atherectomy. Coronary perfo-
with SES; 96.4% with PES; 80.8% with BMS; P ¼ 0.01 for ration may also occur in epicardial or septal collateral braches
comparison), despite significantly greater stent number and related to catheter delivery during retrograde procedures. Fur-
length per patient with DES (79). Similarly, the open-label, ther, perforation is not always manifest during the procedure;
multicenter Asian CTO registry reported no significant differ- in one case series study, 45% of 31 tamponade events were
ences in one-year TVR rates of 3.6% and 6.7% for SES (N ¼ 396) diagnosed after the patient had left the catheterization labora-
and PES (N ¼ 526) (76). In a subgroup of patients with three- tory (82). Although infrequent, coronary perforation and the
year follow-up, MACE was significantly lower in the SES group development of cardiac tamponade may have severe clinical
(10.9% SES vs. 16.3% PES; P ¼ 0.03), although rates of TLR were consequences. Among patients developing cardiac tamponade,
statistically similar (7.7% SES vs. 9.5% PES) (79). Recently, these rates of death, emergency surgery, MI, and transfusion
same investigators reported results from a prospective registry occurred in 42%, 39%, 29%, 65% of patients, respectively (82).
of 1149 CTO patients treated with SES (N ¼ 365), PES (N ¼ 482), Aside from early recognition, management of coronary perfo-
zotarolimus-eluting stents (ZES) (N ¼ 154), tacrolimus-eluting ration requires simultaneous action on behalf of multiple
stents (TES) (N ¼ 109), or endothelial progenitor cell (EPC) catheterization laboratory personnel, including (i) prolonged
capture stents (N ¼ 39) (76). At nine months, TLR was signifi- inflation across the perforation with an occluding balloon or
cantly lower with SES compared with ZES, TES, and EPC perfusion balloon catheter, (ii) reversal of anticoagulation, (iii)
stents, but did not statistically differ from PES. In another covered stent placement, emergency surgery or embolization,
nonrandomized comparison of CTO patients treated with SES and/or (iv) pericardiocentesis. In the absence of clinically evi-
(N ¼ 107) and PES (N ¼ 29), statistically significant differences dent perforation, an operator nevertheless should be vigilant if
were observed regarding angiographic restenosis (9.4% with a perforation is suspected. Angiography of the contralateral
SES vs. 28.6% with PES; P < 0.05), although rates of TVR did vessel must be performed to exclude the possibility of extrav-
not statistically vary (3.7% with SES vs. 6.9% with PES) (78). asation via the collaterals. Close monitoring with a pulmonary
Finally, a modest-sized randomized trial comparing SES artery catheter and serial echocardiograms may also be neces-
(N ¼ 60) and PES (N ¼ 58) in CTO PCI also demonstrated no sary. Finally, aside from coronary perforation, additional pro-
significant difference in the eight-month TVR rates of 3.3% and cedural-related complications that may result in MI, include
7.0% in the SES and PES cohorts, respectively (77). thrombus formation, coronary dissection, and side-branch or
The ongoing PRISON III trial is intended to compare collateral occlusion.
clinical and angiographic outcomes among 300 CTO patients
randomized in a 1:1, open-label fashion to treatment with either
SES or ZES (Medtronic Corp., Santa Rosa, California, U.S.) (80). INNOVATIVE CTO TECHNOLOGIES
The primary endpoint is in-segment late lumen loss at eight- In spite of the enthusiasm for DES to substantially reduce
month angiographic follow-up, and secondary clinical end- restenosis and repeat revascularization, CTOs remain the last
points include TLR, target vessel failure, and stent thrombosis. great barrier to initial technical and procedural success.
Although the complexity of lesions treated with DES is likely
to increase, interventionalists must first be able to recanalize
CTO COMPLICATIONS CTOs before even considering which type of stent to implant.
CTO PCI has traditionally been considered benign, with the Accordingly, several novel technologies have been proposed to
presupposition that because the artery is already occluded with facilitate CTO PCI, with variable success.
collateral circulation, no harm can be done. However, observa- Extending the platform of wire-based technology further
tional studies demonstrate that CTO PCI carries much the same is the Safe-Cross system (Spectranetics, Colorado Springs, Col-
risk as conventional PCI (81). Coronary perforation is the most orado, U.S.) that employs forward-looking optical coherence
feared complication of CTO intervention. Perforation often may reflectometry coupled with a radiofrequency energy source at
[2/7/010/9:18:30] [453–468]
the distal end of the guidewire. In addition to direct visualiza- extended intracoronary infusions of fibrinolytic agents and
tion, optical coherence reflectometry enables guidance within collagenase. Recent preclinical investigation with the local
the CTO, informing and/or alerting the operator when the wire administration of collagenase has suggested that this matrix
engages the intraluminal CTO segment or the vessel wall, metalloproteinase may effectively degrade occlusive fibrotic
respectively. This mechanism of indirect imaging within the plaque to facilitate crossing with standard guidewires (88).
CTO may then facilitate guidewire advancement using the Finally, for patients with CTOs that are not amenable to
energy source to ablate fibrous or calcified tissue. Recent revascularization, performance of transmyocardial laser revas-
evaluations in both coronary and peripheral interventions cularization does not appear to improve survival or general
have demonstrated the potential of this technology, particularly health status (89).
following guidewire failures. Among 116 patients enrolled in
the Guided Radiofrequency Energy Ablation of Total
Occlusions (GREAT) registry, technical success (defined as FUTURE DIRECTIONS
guidewire placement in the true lumen distal to the CTO) Although the angiographic results of intended revasculariza-
was achieved in 56% of patients following conventional guide- tion are unmistakable (i.e., either failed or successful epicardial
wire failure (83). One device-related clinical perforation was recanalization), the effects of total occlusion and reperfusion at
reported. A similar success rate was observed in peripheral the level of the myocardium are less apparent. To identify
arterial occlusions in the Guided Radiofrequency in Peripheral which patients might benefit from revascularization, delayed-
Total Occlusion (GRIP) trial, in which the Safe-Cross system enhancement contrast magnetic resonance imaging (MRI) may
was capable of traversing 76% of CTOs (75 lesions) after failure be useful for the identification of viable and ischemic myocar-
with standard guidewires (75). dium subtended by a CTO. Recent clinical experience has been
An alternative to wire-based technologies is the Front- useful in identifying viable myocardial tissue in spite of
runner catheter (Cordis Corporation, Warren, New Jersey, matched, regional wall motion abnormalities by other imaging
U.S.). Intended to cross through fibrous, resistant CTOs, the methods (90). Among 44 patients with 58 CTO segments, 37 pati-
Frontrunner catheter performs blunt microdissection of tissue ents (64%) had 50% transmural infarction, with 12 patients
within the CTO to facilitate guidewire placement in the distal (21%) having no evidence of infarction. Further, the presence of
true lumen. Specifically, the actuation of jaws on the distal end collateral flow did not predict either myocardial viability or
of a 0.039-in. diameter catheter creates a 2.3-mm excursion that improvement following revascularization. Territories without
separates tissue planes within the occluded segment. The cath- extensive infarction at baseline demonstrated significant regional
eter is approved for use in both coronary and peripheral wall motion improvement following revascularization, and more
arterial interventions and may also be used with a support recent evaluations with adenosine stress MRI following percuta-
catheter intended for greater stability at the proximal occlusion. neous revascularization have demonstrated resolution of ische-
Among 105 patients with CTOs, a previous design of the mia. Thus, application of noninvasive imaging with coronary CT
Frontrunner catheter successfully crossed 56% of CTOs other- angiography and MRI may inform patient selection by assisting
wise impenetrable to conventional guidewires (84). in prediction of both procedural success (coronary CT angiog-
Innovative technologies presently under investigation raphy) and clinical improvement (MRI).
include true lumen crossing and intraluminal reentry devices. Novel methods for CTO revascularization may also look
Using the facilitated antegrade steering technique (FAST), both beyond the occluded vessel segment as an alternative to
the CrossBoss CTO catheter and the Stingray CTO reentry guidewire-based techniques. Although still a developing tech-
system (Bridgepoint Technologies, Minneapolis, Minnesota, nology, percutaneous in situ coronary artery bypass (PICAB)
U.S.) may provide increased success rates and safety compared may provide an alternative method of revascularization for
with guidewires alone. The CrossBoss CTO catheter tracks over patients in whom conventional revascularization has either
a wire via a FAST spin technique. Specifically, the catheter has a failed or is not feasible (91). Specifically, percutaneous in situ
highly torqueable coiled-wire shaft, and the spin technique coronary venous arterialization (PICVA) utilizes the native
reduces the amount of forward force required to cross the coronary vein as a bypass conduit. Using IVUS guidance, an
CTO. The Stingray system is a balloon catheter with a flat adjacent coronary vein is accessed with a nitinol needle that
shape and opposing exit ports for selective guidewire reentry permits passage of a 0.014-in. wire followed by placement of
when positioned within a dissection plane but adjacent to the venous occluder devices proximal and distal to the bypass
vessel true lumen. The first-in-human experience reported a segment. Feasibility studies are ongoing to overcome chal-
technical success of 88% (36/41) with no in-hospital or 30-day lenges in venous anatomy, variable quality of imaging with
MACE (85). The ongoing FAST-CTO clinical trial will ultimately IVUS, and accurate sizing of venous occluder devices. Aside
address the feasibility and effectiveness of this new technology. from venous arterialization, the transvascular access catheter
For many CTO-specific device therapies, clinical success may also be effective in facilitating guidewire reentry from a
has not exceeded expectations compared with historical stan- dissection plane distal to an occluded segment.
dards. In the Total Occlusion Trial with Angioplasty by Laser
Guidewire (TOTAL) trial, no significant differences were
observed between patients treated with standard guidewires CONCLUSIONS
or the excimer laser guidewire [76 (52%) laser vs. 75 (47%) The implications of improving early procedural and long-term
conventional guidewire, P ¼ 0.33] (22). Limited evaluations clinical outcomes among patients with CTOs are considerable.
with catheter-based delivery of high-frequency mechanical However, our current knowledge of procedural, angiographic,
ultrasound have also reported lower procedural success com- and clinical outcomes following PCI for CTOs is limited by the
pared with guidewires (86,87), although investigation with a systematic or preferential exclusion of patients with CTO target
revised catheter platform is planned. Further, pharmacological lesions from major interventional cardiology clinical trials.
approaches to CTO dissolution have been developed with Further, no technology is presently available to reliably predict
[2/7/010/9:18:30] [453–468]
procedural success in patients considered for CTO revascula- occlusion and short and long occluded segments. J Am Coll
rization. Even following successful guidewire crossing, com- Cardiol 1993; 21:604–611.
pared with revascularization in nonocclusive lesions, the 11. Srivatsa SS, Edwards WD, Boos CM, et al. Histologic correlates of
unfavorable results with coronary stenting reflect the complex- angiographic chronic total coronary artery occlusions: influence of
occlusion duration on neovascular channel patterns and intimal
ity of CTOs with regard to lesion length, plaque burden, neg-
plaque composition. J Am Coll Cardiol 1997; 29:955–963.
ative vascular remodeling, thrombus, and calcification. 12. Hosoda Y, Kawano K, Yamasawa F, et al. Age-dependent changes
While the challenges associated with CTOs may seem of collagen and elastin content in human aorta and pulmonary
formidable, it is equally clear that our evolving understanding artery. Angiology 1984; 35:615–621.
of the benefits of revascularization has contributed to increas- 13. Katsuda S, Okada Y, Minamoto T, et al. Collagens in human
ing interest in CTOs, reflected in an increasing number of CTO atherosclerosis. Immunohistochemical analysis using collagen
procedures, the development of novel technologies, and the type-specific antibodies. Arterioscler Thromb 1992; 12:494–502.
design of trials dedicated to CTO revascularization. Over the 14. Burke AP, Kolodgie FD, Farb A, et al. Morphological predictors of
past five years alone, a number of alternative technologies to arterial remodeling in coronary atherosclerosis. Circulation 2002;
guidewires have been advanced, including mechanical techni- 105:297–303.
15. Sueishi K, Yonemitsu Y, Nakagawa K, et al. Atherosclerosis and
ques, ablation, microdissection, and transvascular reentry. Fur-
angiogenesis. Its pathophysiological significance in humans as well
ther clinical trials with DES and drug-eluting balloons are also as in an animal model induced by the gene transfer of vascular
underway. Recent evaluations with MRI in patients with CTOs endothelial growth factor. Ann N Y Acad Sci 1997; 811:311–22; 22–24.
provide further mechanistic support toward improvement in 16. Kumamoto M, Nakashima Y, Sueishi K. Intimal neovasculariza-
left ventricular function and how these findings may translate tion in human coronary atherosclerosis: its origin and pathophy-
into observations of favorable clinical outcomes associated with siological significance. Hum Pathol 1995; 26:450–456.
successful CTO revascularization. Thus, while CTOs reflect the 17. Sakuda H, Nakashima Y, Kuriyama S, et al. Media conditioned by
current inadequacies in PCI methods for achieving initial pro- smooth muscle cells cultured in a variety of hypoxic environments
cedural success and for sustaining restenosis-free patency fol- stimulates in vitro angiogenesis. A relationship to transforming
lowing initial success, they also represent a very unique growth factor-beta 1. Am J Pathol 1992; 141:1507–1516.
18. Puma JA, Sketch MH Jr., Thompson TD, et al. Support for the
opportunity to conquer perhaps the most difficult and yet one
open-artery hypothesis in survivors of acute myocardial infarc-
of the most common lesion subsets in interventional cardiology. tion: analysis of 11,228 patients treated with thrombolytic therapy.
Am J Cardiol 1999; 83:482–487.
19. Werner GS, Surber R, Kuethe F, et al. Collaterals and the recovery
REFERENCES of left ventricular function after recanalization of a chronic total
1. Srinivas VS, Brooks MM, Detre KM, et al. Contemporary percuta- coronary occlusion. Am Heart J 2005; 149:129–137.
neous coronary intervention versus balloon angioplasty for multi- 20. Safley DM, House JA, Marso SP, et al. Improvement in survival
vessel coronary artery disease: a comparison of the National following successful percutaneous coronary intervention of coro-
Heart, Lung and Blood Institute Dynamic Registry and the Bypass nary chronic total occlusions: variability by target vessel. JACC
Angioplasty Revascularization Investigation (BARI) study. Circu- Cardiovasc Interv 2008; 1:295–302.
lation 2002; 106:1627–1633. 21. Olivari Z, Rubartelli P, Piscione F, et al. Immediate results and
2. King SB III, Lembo NJ, Weintraub WS, et al. A randomized trial one-year clinical outcome after percutaneous coronary interven-
comparing coronary angioplasty with coronary bypass surgery. tions in chronic total occlusions: data from a multicenter, prospec-
Emory Angioplasty versus Surgery Trial (EAST). N Engl J Med tive, observational study (TOAST-GISE). J Am Coll Cardiol 2003;
1994; 331:1044–1150. 41:1672–1678.
3. Serruys PW, van Geuns RJ. Arguments for recanalization of 22. Serruys PW, Hamburger JN, Koolen JJ, et al. Total occlusion trial
chronic total occlusions. JACC Cardiovasc Interv 2008; 1:54–55. with angioplasty by using laser guidewire. The TOTAL trial. Eur
4. Bell MR, Berger PB, Bresnahan JF, et al. Initial and long-term Heart J 2000; 21:1797–1805.
outcome of 354 patients after coronary balloon angioplasty of total 23. Rubartelli P, Verna E, Niccoli L, et al. Coronary stent implantation
coronary artery occlusions. Circulation 1992; 85:1003–1011. is superior to balloon angioplasty for chronic coronary occlusions:
5. Noguchi T, Miyazaki MS, Morii I, et al. Percutaneous transluminal six-year clinical follow-up of the GISSOC trial. J Am Coll Cardiol
coronary angioplasty of chronic total occlusions. Determinants of 2003; 41:1488–1492.
primary success and long-term clinical outcome. Catheter Cardi- 24. He ZX, Mahmarian JJ, Verani MS. Myocardial perfusion in
ovasc Interv 2000; 49:258–264. patients with total occlusion of a single coronary artery with
6. Werner GS, Emig U, Mutschke O, et al. Regression of collateral and without collateral circulation. J Nucl Cardiol 2001; 8:452–457.
function after recanalization of chronic total coronary occlusions: a 25. Aboul-Enein F, Kar S, Hayes SW, et al. Influence of angiographic
serial assessment by intracoronary pressure and Doppler record- collateral circulation on myocardial perfusion in patients with
ings. Circulation 2003; 108:2877–2882. chronic total occlusion of a single coronary artery and no prior
7. Tamai H, Berger PB, Tsuchikane E, et al. Frequency and time myocardial infarction. J Nucl Med 2004; 45:950–955.
course of reocclusion and restenosis in coronary artery occlusions 26. Cheng AS, Selvanayagam JB, Jerosch-Herold M, et al. Percuta-
after balloon angioplasty versus Wiktor stent implantation. Am neous treatment of chronic total coronary occlusions improves
Heart J 2004; 147:E9. regional hyperemic myocardial blood flow and contractility:
8. Stone GW, Kandzari DE, Mehran R, et al. Percutaneous recanal- insights from quantitative cardiovascular magnetic resonance
ization of chronically occluded coronary arteries: a consensus imaging. JACC Cardiovasc Interv 2008; 1:44–53.
document: part I. Circulation 2005; 112:2364–2372. 27. Hochman JS, Lamas GA, Buller CE, et al. Coronary intervention
9. Zidar FJ, Kaplan BM, O’Neill WW, et al. Prospective, randomized for persistent occlusion after myocardial infarction. N Engl J Med
trial of prolonged intracoronary urokinase infusion for chronic 2006; 355:2395–2407.
total occlusions in native coronary arteries. J Am Coll Cardiol 28. Dzavik V, Buller CE, Lamas GA, et al. Randomized trial of
1996; 27:1406–1412. percutaneous coronary intervention for subacute infarct-related
10. Katsuragawa M, Fujiwara H, Miyamae M, et al. Histologic studies coronary artery occlusion to achieve long-term patency and
in percutaneous transluminal coronary angioplasty for chronic improve ventricular function: the Total Occlusion Study of
total occlusion: comparison of tapering and abrupt types of Canada (TOSCA)-2 trial. Circulation 2006; 114:2449–2457.
[2/7/010/9:18:30] [453–468]
29. Abbate A, Biondi-Zoccai GG, Appleton DL, et al. Survival and 48. Sirnes PA, Myreng Y, Molstad P, et al. Improvement in left ven-
cardiac remodeling benefits in patients undergoing late percuta- tricular ejection fraction and wall motion after successful recanaliza-
neous coronary intervention of the infarct-related artery: evidence tion of chronic coronary occlusions. Eur Heart J 1998; 19:273–281.
from a meta-analysis of randomized controlled trials. J Am Coll 49. Sallam M, Spanos V, Briguori C, et al. Predictors of re-occlusion
Cardiol 2008; 51:956–964. after successful recanalization of chronic total occlusion. J Invasive
30. Stone GW, Reifart NJ, Moussa I, et al. Percutaneous recanalization Cardiol 2001; 13:511–515.
of chronically occluded coronary arteries: a consensus document: 50. Werner GS, Bahrmann P, Mutschke O, et al. Determinants of
part II. Circulation 2005; 112:2530–2537. target vessel failure in chronic total coronary occlusions after
31. Mehran R. Analysis of the Columbia-Milan CTO Registry. stent implantation. The influence of collateral function and coro-
Presented at: The Sixth International Chronic Total Occlusion nary hemodynamics. J Am Coll Cardiol 2003; 42:219–225.
Summit, New York, 2009. 51. Suttorp MJ, Laarman GJ, Rahel BM, et al. Primary Stenting of
32. Mollet NR, Hoye A, Lemos PA, et al. Value of preprocedure Totally Occluded Native Coronary Arteries II (PRISON II): a
multislice computed tomographic coronary angiography to pre- randomized comparison of bare metal stent implantation with
dict the outcome of percutaneous recanalization of chronic total sirolimus-eluting stent implantation for the treatment of total
occlusions. Am J Cardiol 2005; 95:240–243. coronary occlusions. Circulation 2006; 114:921–928.
33. Colombo A, Mikhail GW, Michev I, et al. Treating chronic total 52. Serruys P. Syntax Trial: Chronic total occlusion subsets, 1 year
occlusions using subintimal tracking and reentry: the STAR tech- results from the SYNTAX Trial. Presented at: Cardiovascular
nique. Catheter Cardiovasc Interv 2005; 64:407–411; discussion 12. Research Technologies, Washington, D.C., 2009.
34. Surmely JF, Katoh O, Tsuchikane E, et al. Coronary septal collat- 53. Holmes D. Complex lesions in the e-Cypher Registry. Presented at:
erals as an access for the retrograde approach in the percutaneous Transcatheter Cardiovascular Therapeutics, Washington, D.C., 2004.
treatment of coronary chronic total occlusions. Catheter Cardio- 54. Hoye A, Tanabe K, Lemos PA, et al. Significant reduction in
vasc Interv 2007; 69:826–832. restenosis after the use of sirolimus-eluting stents in the treatment
35. Surmely JF, Tsuchikane E, Katoh O, et al. New concept for CTO of chronic total occlusions. J Am Coll Cardiol 2004; 43:1954–1958.
recanalization using controlled antegrade and retrograde subintimal 55. Werner GS, Krack A, Schwarz G, et al. Prevention of lesion
tracking: the CART technique. J Invasive Cardiol 2006; 18:334–338. recurrence in chronic total coronary occlusions by paclitaxel-
36. Sirnes PA, Golf S, Myreng Y, et al. Stenting in Chronic Coronary eluting stents. J Am Coll Cardiol 2004; 44:2301–2306.
Occlusion (SICCO): a randomized, controlled trial of adding stent 56. Nakamura S, Muthusamy TS, Bae JH, et al. Impact of sirolimus-
implantation after successful angioplasty. J Am Coll Cardiol 1996; eluting stent on the outcome of patients with chronic total occlu-
28:1444–1451. sions. Am J Cardiol 2005; 95:161–166.
37. Engelstein E, Terres W, Hofmann D, et al. Improved global and 57. Ge L, Iakovou I, Cosgrave J, et al. Immediate and mid-term
regional left ventricular function after angioplasty for chronic outcomes of sirolimus-eluting stent implantation for chronic
coronary occlusion. Clin Investig 1994; 72:442–447. total occlusions. Eur Heart J 2005; 26:1056–1062.
38. Mori M, Kurogane H, Hayashi T, et al. Comparison of results of 58. Abizaid A, Chan C, Lim YT, et al. Twelve-month outcomes with a
intracoronary implantation of the Plamaz-Schatz stent with con- paclitaxel-eluting stent transitioning from controlled trials to
ventional balloon angioplasty in chronic total coronary arterial clinical practice (the WISDOM Registry). Am J Cardiol 2006;
occlusion. Am J Cardiol 1996; 78:985–989. 98:1028–1032.
39. Hoher M, Wohrle J, Grebe OC, et al. A randomized trial of elective 59. Grube E, Biondi-Zoccai G, Sangiorgi G, et al. Assessing the safety and
stenting after balloon recanalization of chronic total occlusions. J effectiveness of TAXUS in 183 patients with chronic total occlusions:
Am Coll Cardiol 1999; 34:722–729. insights from the TRUE study. Am J Cardiol 2005; 96:37H.
40. Buller CE, Dzavik V, Carere RG, et al. Primary stenting versus 60. Buellesfeld L, Gerckens U, Mueller R, et al. Polymer-based pacli-
balloon angioplasty in occluded coronary arteries: the Total Occlu- taxel-eluting stent for treatment of chronic total occlusions of
sion Study of Canada (TOSCA). Circulation 1999; 100:236–242. native coronaries: results of a Taxus CTO registry. Catheter
41. Lotan C, Rozenman Y, Hendler A, et al. Stents in total occlusion Cardiovasc Interv 2005; 66:173–177.
for restenosis prevention. The multicentre randomized STOP 61. Kandzari DE, Rao SV, Moses JW, et al. Clinical and angiographic
study. The Israeli Working Group for Interventional Cardiology. outcomes with sirolimus-eluting stents in total coronary occlu-
Eur Heart J 2000; 21:1960–1966. sions: the ACROSS/TOSCA-4 (Approaches to Chronic Occlusions
42. Sirnes PA, Golf S, Myreng Y, et al. Sustained benefit of stenting With Sirolimus-Eluting Stents/Total Occlusion Study of Coronary
chronic coronary occlusion: long-term clinical follow-up of the Arteries-4) trial. JACC Cardiovasc Interv 2009; 2:97–106.
Stenting in Chronic Coronary Occlusion (SICCO) study. J Am Coll 62. Nakamura S, Selvan TS, Bae JH, et al. Impact of sirolimus-eluting
Cardiol 1998; 32:305–310. stents on the outcome of patients with chronic total occlusions:
43. Serruys PW, de Jaegere P, Kiemeneij F, et al. A comparison of multicenter registry in Asia. J Am Coll Cardiol 2003; 43:35A.
balloon-expandable-stent implantation with balloon angioplasty 63. Schomig A, Dibra A, Windecker S, et al. A meta-analysis of 16
in patients with coronary artery disease. Benestent Study Group. randomized trials of sirolimus-eluting stents versus paclitaxel-
N Engl J Med 1994; 331:489–495. eluting stents in patients with coronary artery disease. J Am Coll
44. Serruys PW, van Hout B, Bonnier H, et al. Randomised comparison Cardiol 2007; 50:1373–1380.
of implantation of heparin-coated stents with balloon angioplasty 64. Nakamura S, Bae JH, Muthusamy TS, et al. Four-year durability of
in selected patients with coronary artery disease (Benestent II). sirolimus-eluting stents in patients with chronic total occlusions
Lancet 1998; 352:673–681. compared with bare metal stents: multicenter registry in Asia. Am
45. Buller CE, Teo KK, Carere RG, et al. Three year clinical outcomes J Cardiol 2007; 100:93L.
from the Total Occlusion Study of Canada (TOSCA). Circulation 65. Dibra A, Kastrati A, Mehilli J, et al. Paclitaxel-eluting or sirolimus-
2000; 102:II1885. eluting stents to prevent restenosis in diabetic patients. N Engl J
46. Dzavik V, Carere RG, Mancini GB, et al. Predictors of improvement Med 2005; 353:663–670.
in left ventricular function after percutaneous revascularization of 66. Goy JJ, Stauffer JC, Siegenthaler M, et al. A prospective random-
occluded coronary arteries: a report from the Total Occlusion Study ized comparison between paclitaxel and sirolimus stents in the
of Canada (TOSCA). Am Heart J 2001; 142:301–308. real world of interventional cardiology: the TAXi trial. J Am Coll
47. Melchior JP, Doriot PA, Chatelain P, et al. Improvement of left Cardiol 2005; 45:308–311.
ventricular contraction and relaxation synchronism after recanal- 67. Kastrati A, Dibra A, Eberle S, et al. Sirolimus-eluting stents vs
ization of chronic total coronary occlusion by angioplasty. J Am paclitaxel-eluting stents in patients with coronary artery disease:
Coll Cardiol 1987; 9:763–768. meta-analysis of randomized trials. JAMA 2005; 294:819–825.
[2/7/010/9:18:30] [453–468]
68. Kastrati A, Mehilli J, von Beckerath N, et al. Sirolimus-eluting 80. Suttorp MJ, Laarman GJ. A randomized comparison of sirolimus-
stent or paclitaxel-eluting stent vs balloon angioplasty for preven- eluting stent implantation with zotarolimus-eluting stent implan-
tion of recurrences in patients with coronary in-stent restenosis: a tation for the treatment of total coronary occlusions: rationale and
randomized controlled trial. JAMA 2005; 293:165–171. design of the PRImary Stenting of Occluded Native coronary
69. Morice MC, Colombo A, Meier B, et al. Sirolimus- vs paclitaxel- arteries III (PRISON III) study. Am Heart J 2007; 154:432–435.
eluting stents in de novo coronary artery lesions: the REALITY 81. Suero JA, Marso SP, Jones PG, et al. Procedural outcomes and
trial: a randomized controlled trial. JAMA 2006; 295:895–904. long-term survival among patients undergoing percutaneous cor-
70. Stettler C, Wandel S, Allemann S, et al. Outcomes associated with onary intervention of a chronic total occlusion in native coronary
drug-eluting and bare-metal stents: a collaborative network meta- arteries: a 20-year experience. J Am Coll Cardiol 2001; 38:409–414.
analysis. Lancet 2007; 370:937–948. 82. Fejka M, Dixon SR, Safian RD, et al. Diagnosis, management, and
71. Windecker S, Remondino A, Eberli FR, et al. Sirolimus-eluting and clinical outcome of cardiac tamponade complicating percutaneous
paclitaxel-eluting stents for coronary revascularization. N Engl J coronary intervention. Am J Cardiol 2002; 90:1183–1186.
Med 2005; 353:653–662. 83. Baim DS, Braden G, Heuser R, et al. Utility of the Safe-Cross-
72. de Lezo JS, Medina A, Pan M, et al. Drug-eluting stents for complex guided radiofrequency total occlusion crossing system in chronic
lesions: latest angiographic data from the randomized rapamycin coronary total occlusions (results from the Guided Radio Fre-
versus paclitaxel CORPAL study J Am Coll Cardiol 2005; 45:75A. quency Energy Ablation of Total Occlusions Registry Study). Am J
73. Hoye A, Ong ATL, Aoki J, et al. Drug-eluting stent implantation Cardiol 2004; 94:853–858.
for chronic total occlusions: comparison between the sirolimus- 84. Chadwick D. The Frontrunner CTO catheter experience. Cath Lab
and paclitaxel-eluting stent. EuroIntervention 2005; 1:193–197. Digest 2002:10.
74. Nakamura S, Bae JH, Cahyadi YH, et al. Comparison of efficacy 85. Available at: http://www.BridgePointMedical.com. Accessed
and safety between sirolimus-eluting stent and paclitaxel-eluting June 3, 2009.
stent on the outcome of patients with chronic total occlusions: 86. Cannon LA, John J, LaLonde J. Therapeutic ultrasound for chronic
multicenter registry in Asia. Am J Cardiol 2005; 96:38H. total coronary artery occlusions. Echocardiography 2001; 18:219–223.
75. Wholey MH. Recanalization of peripheral total occlusions. Endo- 87. Cannon L, Siegel R, Greenberg J, et al. Recanalization of total
vascular Today, 2003; 2:1–4. coronary occlusions using the Sonicross low frequency ultrasound
76. Nakamura S, Bae JH, Yeo HC, et al. Drug-eluting stents for the catheter. J Am Coll Cardiol 2000; 35:41A.
treatment of chronic total occlusion: a comparison of sirolimus, 88. Strauss BH, Goldman L, Qiang B, et al. Collagenase plaque
paclitaxel, zotarolimus, tacrolimus-eluting and EPC capture digestion for facilitating guidewire crossing in chronic total occlu-
stents: multicenter registry in Asia. Am J Cardiol 2007; 100:16L. sions. Circulation 2003; 108:1259–1262.
77. de Lezo JS, Medina A, Pan M, et al. Drug-eluting stents for the 89. Stone GW, Teirstein PS, Rubenstein R, et al. A prospective, multi-
treatment of chronic total occlusions: a randomized comparison center, randomized trial of percutaneous transmyocardial laser
of rapamycin- versus paclitaxel-eluting stents. Circulation 2005; revascularization in patients with nonrecanalizable chronic total
112:II477. occlusions. J Am Coll Cardiol 2002; 39:1581–1587.
78. Jang JS, Hong MK, Lee CW, et al. Comparison between sirolimus- 90. Kim HW, Shah D, Patel M, et al. Assessment of viability in
and paclitaxel-eluting stents for the treatment of chronic total patients with chronic total occlusions. Circulation 2003; 108:IV698.
occlusions. J Invasive Cardiol 2006; 18:205–208. 91. Oesterle SN, Reifart N, Hayase M, et al. Catheter-based coronary
79. Nakamura S, Bae JH, Cahyadi YH, et al. Comparison of efficacy bypass: a development update. Catheter Cardiovasc Interv 2003;
and durability of sirolimus-eluting stents and paclitaxel-eluting 58:212–218.
stents in patients with chronic total occlusions: multicenter regis-
try. Am J Cardiol 2007; 100:93L.
[2/7/010/9:19:38] [469–481]
36
Saphenous vein grafts

Claudia P. Hochberg and Joseph P. Carrozza
INTRODUCTION compromise, or may be clinically silent, especially if the grafted

In 1969, Favalaro described the first use of coronary artery artery subtends a minimal amount of viable myocardium.
bypass surgery (CABG) for the treatment of myocardial ische-
mia due to coronary artery disease (1). Subsequent advances in
myocardial preservation and surgical techniques led to Pathology of SVG Failure
improvements in outcomes and the widespread use of CABG. Within one year following CABG, most SVG have developed
Whereas CABG improves long-term prognosis in high-risk some degree of concentric intimal proliferation similar to the
subsets, such as patients with significant left main and three- smooth muscle cell proliferative response observed after vascu-
vessel disease with decreased left ventricular function, for most lar injury induced by percutaneous coronary intervention (PCI).
patients CABG is a palliative, rather than a curative procedure. The development of intimal hyperplasia in “arterialized” veins is
The use of the internal mammary artery (IMA) and other a ubiquitous process commencing approximately four weeks
arterial conduits revolutionized CABG because of the relative after implantation. While this proliferative response often results
resistance of arterial conduits to accelerated atherosclerosis in mild luminal narrowing throughout the SVG, it may be of
compared with saphenous vein bypass grafts (SVG) (2,3). sufficient magnitude to obstruct blood flow. The etiology of this
Recurrent ischemic events occur in a time-dependent manner pathologic response probably involves endothelial cell damage
and may result from progression of disease in the native or denudation occurring during harvesting, vascular ischemia
coronary arteries. However, the most common cause of adverse when vasa vasorum are severed, and barotrauma resulting from
cardiac events in the post-CABG patient is atherosclerosis and exposure of the vein to arterial pressure. This proliferative
attrition of SVG (4). response is often most exuberant at anastomotic sites suggesting
greater injury at these loci. Even if intimal hyperplasia does not
result in ischemia within the first year following CABG, its
ANATOMIC CONSIDERATIONS presence may provide the substrate for accelerated SVG athero-
SVG Patency and Predictors of SVG Failure sclerosis (5). Caňos et al. compared the clinical, angiographic,
Ischemia in myocardial territories supplied by SVG is often and intravascular ultrasound (IVUS) characteristics of early
highly dependent on the time from implantation. Historically, (mean SVG age 6 months) and late (mean SVG age 105 months)
the perioperative SVG occlusion rate was 10% and the 10-year SVG failure. Early SVG failures were more likely to be ostial
occlusion rate was 50% (Fig. 36.1) (5). A more recent assessment (37.5% vs. 17.5%, p < 0.001), and had smaller pretreatment
reported improved 10-year SVG patency rates, but lower IMA reference diameters (2.47 0.86 vs. 3.26 0.83, p < 0.001),
patency rates than had previously been reported (6). At 7 to smaller minimum lumen diameters (MLD) (0.80 0.64 vs.
10 days, 95% of SVG were patent by angiography. One week 1.08 0.64, p < 0.001), and greater pretreatment diameter
patency was associated with a 76% 6-year patency rate and 68% stenosis (71.6 19% vs. 66.7 17.7%, p < 0.017). Additionally,
10-year patency (Fig. 36.2). The overall 10-year SVG patency rate early SVG failure was associated with lower thrombolysis in
was 61%. IMA patency at one week was 99% and was associated myocardial infarction (TIMI) flow rates and SVG that failed late
with 90% 6-year and 88% 10-year patency rates. The overall 10- were more likely to have degenerated, diffuse atheroma. IVUS
year IMA patency rate was 85% (6). analysis suggested that SVG that fail early are diffusely diseased
Early SVG failure may result from any process that conduits without positive remodeling, resulting in smaller
contributes to flow reduction and thrombosis such as pro- lumen size (7). While it is controversial as to whether SVG
thrombotic states, surgical misadventure (e.g., constrictive undergo positive remodeling, the lack of positive remodeling
sutures), nonlaminar flow patterns secondary to SVG-coronary may contribute to early graft failure.
artery size mismatch, compromised outflow from distal coro- Premature atherosclerosis is the major factor contribu-
nary artery disease, and venous varicosities. Predictors of SVG ting to the SVG attrition and accounts for the majority of acute
patency include the location of the SVG, with left anterior ischemic syndromes in the post-CABG patient. While the
descending artery SVG having a 69% 10-year patency rate pathophysiology of atherosclerosis in SVG generally resem-
compared with 56% for right coronary artery SVG, and 58% bles that of native coronary arteries, there are important
for left circumflex artery SVG. The diameter of the recipient differences. Compared with native arteries, SVG atheroscle-
artery by angiography was also highly predictive. Grafting of rosis is often rapidly progressive, more diffuse, associated
arteries >2.0 mm in diameter had an 88% 10-year patency rate with greater numbers of foam and inflammatory cells,
compared with arteries 2.0 mm, which had a 55% 10-year lacks a fibrous cap, and is generally more friable (Fig. 36.3).
patency rate (p < 0.001) (6). Clinical presentations of early SVG Furthermore, since SVG lack side branches, any process
failure can be protean and include exertional chest pain, acute resulting in decreased flow may precipitate secondary throm-
coronary syndromes, ventricular arrhythmias, hemodynamic bosis. Arterial bypass conduits are relatively resistant to
[2/7/010/9:19:38] [469–481]
Figure 36.1 A 50-years-old man underwent coronary artery

bypass surgery. Three days later, he developed recurrent chest
pain and ST segment elevation. Angiography revealed a subtotal Figure 36.3 A 20-years-old saphenous vein graft with diffuse
occlusion at the aorto-ostial anastomosis of the vein graft to the atheromatous degeneration.
posterior descending artery.
FUNDAMENTALS
Percutaneous SVG Intervention: the “Prestent Era”
Compared with the treatment of native coronary arteries, balloon
angioplasty of SVG lesions is associated with poorer acute out-
comes as the friable nature of SVG atheroma and greater plaque
burden make balloon dilatation of a lesion less predictable (8).
Angioplasty of SVG that are older, degenerated, diffusely dis-
eased, or totally occluded is associated with a markedly higher
incidence of procedural complications including a high fre-
quency of creatine kinase-MB (CK-MB) elevations (9). In addi-
tion, the long-term clinical and angiographic results of balloon
angioplasty are limited by an incidence of angiographic and
clinical restenosis that may exceed 50% for lesions located at the
ostium or within the body of the graft (10). One large series
reported a five-year eventfree survival rate of only 26% for
patients treated with balloon angioplasty for SVG lesions (11).
The use of debulking techniques has also been associated with
suboptimal outcomes. Directional or extraction atherectomy has
been limited by a high incidence of distal embolization, and
excimer laser angioplasty has been associated with high residual
stenoses with rates of restenosis exceeding 50% (12–14).
SVG Stenting: Early Experience

Figure 36.2 Plot of time-related graft patency (or freedom from The favorable angiographic and clinical outcomes observed after
graft occlusion) for SVG and IMA grafts. p < 0.001 (IMA vs. SVG) stenting native coronary arteries provided a rationale for inves-
(*). Abbreviations: SVG, saphenous vein graft; IMA, internal mam- tigation into the use of stents for the treatment of SVG disease.
mary artery; CABG, coronary artery bypass grafting. Source: From
Several single and multicenter observational studies reported on
Ref. 6.
the use of the Palmaz-Schatz coronary stent (15–17). The multi-
center Palmaz-Schatz Stent Registry enrolled 589 patients with
624 focal SVG lesions. Procedural success was high (>98%) and
major adverse cardiac events such as myocardial infarction (MI),
atherosclerosis, thus offering a greater likelihood of long-term urgent CABG, and death occurred in only 2.9%. Stent thrombosis
freedom from ischemic events (8). However, focal stenoses within the first month was diagnosed in only 1.4% of patients.
may develop within the first year, especially at anastomotic However, there was a high incidence of hemorrhagic complica-
sites. Repeat CABG is technically more challenging than the tions (14.3%) secondary to the mandated use of aspirin, heparin,
first operation and is associated with increased procedural dextran, dipyridamole, and warfarin. Quantitative angiographic
morbidity and mortality (9). analysis of a subset of patients in the registry revealed an overall
[2/7/010/9:19:38] [469–481]
SAPHENOUS VEIN GRAFTS 471
restenosis rate of 34% (15), with significantly higher rates of treatment modality for SVG disease. While stand-alone balloon
restenosis observed for previously treated (51% vs. 22%) and angioplasty can be performed with similar safety and efficacy
aorto-ostial (61% vs. 28%) lesions. Debulking of ostial SVG as stenting at the distal anastomosis of SVG, adjunctive stenting
lesions before stent placement did not appear to confer improved improves long-term outcome by reducing the need for repeat
long-term outcomes compared with stenting alone (18). This revascularization (20).
relatively high incidence of restenosis may have been due in
part to the absence of stents designed to optimally treat vessels Bare-Metal Vs. Drug-Eluting Stents in SVG Disease
larger than 4 mm in diameter. On the basis of the findings of the SAVED trial and the more
recent VENESTENT trial, the implantation of stents in diseased
The SAVED Trial SVG has become the standard of care for the endovascular
The first multicenter, prospective randomized trial of SVG treatment of SVG lesions (21). Despite improved angiographic
stenting was the Saphenous Vein in De Novo (SAVED) trial, results and clinical outcomes after bare-metal stent (BMS)
in which 220 patients were randomized to Palmaz-Schatz implantation, the rate of BMS restenosis in SVG is higher than
stenting or balloon angioplasty for the treatment of relatively in the native coronary arteries, with restenosis rates over 30%
focal, de novo lesions in 3- to 5-mm SVG (19). The primary (19). The introduction of drug-eluting stents (DES) has conclu-
endpoint of the trial was angiographic restenosis at six months. sively decreased the rates of angiographic and clinical restenosis,
Procedural success (defined as a reduction in diameter stenosis as well as target vessel revascularization (TVR) in native coro-
<50%, in the absence of major cardiac complications) was nary arteries compared with BMS. Whether there is comparable
significantly higher with stenting compared with balloon benefit to DES implantation in SVG is unknown. Several regis-
angioplasty (92% vs. 69%), although the incidence of major tries and one small randomized trial evaluating both the short-
hemorrhagic complications was also higher in the stent cohort. and the long-term outcomes of DES and BMS in the treatment of
The posttreatment minimal luminal diameter was significantly SVG disease have been completed (22–26). The Reduction of
larger in the stent group (2.81 vs. 2.16 mm) and at six months, Restenosis in Saphenous Vein Grafts with Sirolimus-Eluting
despite a greater late loss (1.06 vs. 0.66 mm), stenting conferred Stents (RRISC) trial randomized 75 patients with SVG lesions
both a significantly greater net gain (0.85 vs. 0.54 mm) and (distal graft anastomotic lesions were excluded) to receive BMS
larger minimal luminal diameter (1.73 vs. 1.49 mm). Whereas or sirolimus DES (22). In-stent late loss was significantly lower in
the angiographic restenosis rates were not statistically different DES (0.38 0.51 vs. 0.79 0.66 mm) compared with BMS, and
(37% vs. 46%, p ¼ 0.11) because of inadequate sample size, both binary in-stent and in-segment restenosis were reduced
freedom from major adverse cardiac events [MACE defined as (11.3% vs. 30.6% and 13.6% vs. 32.6%, respectively). Addition-
freedom from death, MI, repeat CABG, or target lesion revas- ally, both TLR and TVR were significantly reduced with DES
cularization (TLR)] was significantly improved in the stent compared with BMS, but there were no significant differences in
group (73% vs. 58%, p ¼ 0.03) (Fig. 36.4). Although the rates of death or MI between the two groups. However, inter-
SAVED trial did not conclusively demonstrate a reduction in mediate term follow-up (median 32 months) reported an excess
angiographic restenosis with stenting, the favorable clinical mortality in the cohort randomized to DES with 11 deaths in the
results from the trial as well as the limitations of other devices, DES group compared with 0 deaths in the BMS group (p < 0.00)
has established stenting as the predominant percutaneous (Fig. 36.5A). A post hoc analysis of the RRISC trial suggests that
the reduction in repeat revascularization in the DES group seen
at six months is lost at later follow-up (Fig. 36.5B) (27). One
possible mechanism of DES failure in SVG was the potential for
stent-SVG size mismatch. Many SVG are >4.0 mm in diameter
and DES are not currently manufactured larger than 3.5 mm.
This potential size mismatch may lead to stent undersizing and
increased rates of restenosis. Another possibility is that although
DES implantation in SVG is associated with a lower early risk of
restenosis, there may be a significant “catch-up phenomenon”
(similar to that seen with brachytherapy) resulting in later
events.
Recent results from the Registro Regionale Angiolastiche
(REAL) Emilia-Romagna Registry from Italy also suggested
similar long-term clinical outcomes of DES compared with
BMS in the treatment of SVG disease (25). The registry con-
tained 288 BMS patients and 72 DES patients. The incidence of
MACE at 12 months was similar in the two groups (17.8% with
DES and 20.3% with BMS, p ¼ 0.460). Nor was there a differ-
ence in the individual components of the primary endpoint.
Figure 36.4 Saphenous Vein in De Novo (SAVED) trial. Kaplan- Additionally the TLR was not significantly different between
Meier survival curves for freedom from major cardiac events. Event- the two groups, althouth there was a two-fold increase in TLR
free survival was significantly higher among patients assigned to in the BMS group compared with the DES group (8.1% vs.
stenting than among those assigned to angioplasty. The relative 4.3%) (25). In contrast, a single center study with 61 DES and 89
risk of a major cardiac event after stenting was 0.82 (95% confidence BMS showed cumulative MACE rates at six months to be 11.5%
interval, 0.68–0.98); after adjustment for diabetes, the relative risk in the DES group compared with 28.1% in BMS group, (p ¼ 0.02)
was 0.82 (95% confidence interval, 0.68–0.99). Source: From Ref. 19. and TLR rates of 3.3% vs. 19.8%, respectively (24). Similarly,
Lee et al. looked at 139 patients undergoing SVG PCI and
[2/7/010/9:19:38] [469–481]
Figure 36.5 (A) Kaplan-Meier estimates for survival by type of randomized stent. (B) Kaplan-Meier estimates for target vessel
revascularization–free survival by type of randomized stent. Abbreviations: BMS, bare-metal stent; SES, sirolimus-eluting stent. Source:
From Ref. 27.
found that DES was associated with a lower incidence of MI acute coronary syndromes, plaque ulceration, and SVG degen-
(4.3% vs. 20%, p ¼ 0.04), and TVR (1.01% vs. 36.9%, p ¼ 0.035) at a eration (33).
mean follow-up of 9.1 months (23). There are even fewer data The inability to restore TIMI 3 flow is associated with a
available on late stent restenosis in DES-treated patients. A large 32% incidence of Q-wave, or large (CK-MB > 50 IU/L) non–Q
randomized trial with long-term follow-up is needed to better wave MI, and an 8% in-hospital mortality, underscoring the
delineate the outcomes of DES compared with BMS in the treat- link between reduced postintervention TIMI flow and clinical
ment of SVG disease. outcome (34). Rates of periprocedural MI in one study were
41.7% with postprocedure TIMI 1 or 2 flow, and only 5.6% with
TIMI 3 flow (26). However, it is important to remember that
COMPLICATIONS OF SVG STENTING even in the absence of angiographic complications or impaired
Distal Embolization and No-Reflow flow, patients who undergo successful SVG stenting are at risk
The major limitation of SVG stenting is the occurrence of acute for periprocedural myonecrosis. The incidence and consequen-
ischemic events secondary to distal embolization and no- ces of distal embolization following SVG stenting are best
reflow. When serial cardiac biomarkers are measured routinely illustrated by a retrospective analysis of 1056 patients by
following SVG stenting, the incidence of periprocedural CK- Hong et al. (28). Seventy percent of the patients did not expe-
MB elevation may be as high as 50%, with an incidence of large rience an angiographic complication, however posttreatment
(>5 times normal) CK-MB release of 15% (28). The pathophysi- CK-MB elevation was still an important determinant of mor-
ology of distal embolization following SVG stenting is multi- tality at 12 months indicating that angiography is inadequate to
factorial. Compared with native coronary arteries, diseased completely evaluate downstream microperfusion. Postproce-
SVG are on average 0.3 mm in diameter larger and more dure enzyme elevation and diabetes mellitus were the strongest
diffusely diseased, and thus contain a greater plaque burden independent predictors of long-term mortality. This strong
(29). Furthermore, the friable nature of a SVG atheroma, with association between periprocedural myonecrosis and poor
its abundance of cholesterol-laden foam cells and thin fibrous prognosis in post-CABG patients (compared with patients
caps renders it prone to embolization following high-pressure undergoing stenting of native ateries) is best explained by the
dilatation commonly employed to optimize stent expansion. observation that this population is older, and has a greater
Aspirates from stented SVG are enriched with atheromatous incidence of comorbidities such as congestive heart failure,
elements, that is, foam cells, inflammatory cells, fibrous caps, diabetes mellitus, and reduced ejection fraction (29). A more
and necrotic cores (30). Older, degenerated SVG often contain recent pooled analysis of five randomized control trials and one
abundant fresh and organized thrombus, which develop in registry study conducted by Coolong et al. evaluating embolic
SVG with reduced or nonlaminar flow resulting from vascular protection devices (EPDs) in SVG PCI evaluated determinants
ectasia, flow-limiting atheroma, poor distal outflow, and sys- of MACE at 30 days. The univariate predictors of increased risk
temic prothrombotic states associated with comorbidities such were current smoking, history of MI, glycoprotein (GP) IIb/IIIa
as diabetes mellitus (31). Disruption of fresh thrombus may inhibitor use, the angiographic presence of thrombus and two
lead to no-reflow from the embolization of platelet aggregates novel angiographic determinants, the SVG degeneration score
or intense microvascular constriction following the release of and the estimated plaque volume (35). The SVG degeneration
soluble mediators of vasoconstriction such as serotonin or score is determined by estimating the degree of luminal irreg-
thromboxane A2 (32). Risk factors for distal embolization ularities or ectasia making up >20% of the reference normal
include angiographically visible thrombus, intervention for segment. Multivariate analysis found that the strongest
[2/7/010/9:19:38] [469–481]
independent predictors of MACE were the SVG degeneration associated with increased MACE risk, potentially because of
score (p < 0.0001) and the estimated plaque volume (p < 0.001). biased patient selection (35). Moreover, a meta-analysis of five
However, the presence of thrombus, increasing patient age, the randomized trials of GP IIb/IIIa receptor antagonists also failed
use of a GP IIb/IIIa inhibitor and current smoking still carried to demonstrate improved outcomes in 627 patients undergoing
lesser independent risk. The use of GP IIb/IIIa inhibitors was SVG PCI (40). At 30 days, death, MI, or TVR occurred in 16.5% of
not randomized in these studies, therefore a detrimental effect patients treated with GP IIb/IIIa receptor antagonists versus
could not be determined. The finding that GP IIb/IIIa inhibitor 12.6% in the placebo group. One small series demonstrated
use was associated with a higher risk of MACE independent of that local delivery of abciximab significantly reduced the lesion
the angiographic predictors may reflect that these medications percent diameter stenosis and TIMI thrombus grade (41).
are chosen for use in patients with a perceived higher prepro- Thrombolytic agents such as urokinase are moderately
cedural risk (35). Likewise, in a post hoc analysis of the PRIDE effective in recanalizing occluded SVG, but are associated with
study, Kirtane found that lesion length was an independent a high incidence (>10%) of hemorrhagic complications (42).
predictor of both CK-MB release and adverse events in patients There has been recent interest in the intracoronary administra-
undergoing SVG PCI with EPDs (36). These studies underscore tion of fibrinolytic agents to help dissolve thrombus in an effort
the unequivocal link between SVG atheroma, procedural to improve microvascular dysfunction post-PCI without requir-
events, and clinical outcome. The use of an EPD offers a 25% ing systemic doses of fibrinolytics. Intragraft administration of
to 40% reduction in 30-day MACE, however the lack of other fibrinolytics has not been studied, but may offer an alternative
effective treatments or strategies to prevent distal embolization approach to treating SVG with large thrombus burden. One
is the impetus for a large body of research focusing on adjunc- approach to reducing the thrombus burden before SVG stenting,
tive pharmacologic and mechanical treatments to render SVG without exposing the patient to the risk of systemic fibrinolysis,
stenting safer and more predictable. is mechanical thrombectomy. The AngiojetTM (Possis Medical,
There is some evidence that direct stenting of SVG (i.e., Inc., Minneapolis, Minnesota, U.S.) utilizes the Venturi principle
without predilation) may be associated with a reduction in the to create a low-pressure vortex around the catheter tip, macerat-
volume of atheroembolic debris (30). Another strategy to ing fresh thrombus before aspiration (Fig. 36.6). In the VEGAS
reduce distal embolization during SVG stenting is to deploy a trial, 351 patients with thrombotic lesions, the majority of which
self-expanding stent such as the Wallstent, without predilation. were SVG, were randomized to prolonged urokinase infusion or
It was hypothesized that primary stenting with these self- thrombectomy with the Angiojet (43). Randomization to Angiojet
expanding endovascular prostheses might reduce the likeli- treatment was associated with greater procedural success,
hood of distal embolization by trapping friable atheroma before reduced incidence of major adverse events, shorter length of
high-pressure balloon dilatation. However, thus far this strat- hospitalization, and reduction in in-hospital cost. However,
egy has not been validated in a prospective clinical trial. despite a significant reduction in angiographic thrombus, 11%
of patients still had periprocedural MI, affirming the growing
understanding that factors beyond thrombosis, for example,
ADJUNCTIVE THERAPY AND EQUIPMENT atheroembolism and soluble mediators are major contributors
Pretreatment with Vasodilators to adverse events after SVG stenting. Thrombus aspiration using
An important mediator of no-reflow is downstream microvas- hollow-lumen catheters such as the ExportTM (Medtronic, Inc.,
cular vasoconstriction. The intragraft administration of a variety Santa Rosa, California, U.S.) have also been used to remove
of microvascular vasodilators such as verapamil, diltiazem, thrombus in SVG prior to stenting.
adenosine, nicardipine, and nitroprusside have been shown to Bivalirudin, a direct thrombin inhibitor, is increasingly
improve flow in no-reflow events (37–39). Prophylactic treatment administered instead of unfractionated heparin as an antico-
of SVG before the intervention with verapamil, and more agulant during PCI of native coronary arteries. It has been
recently with nicardipine, has been shown to decrease the occur- associated with lower rates of bleeding without increased
rence rates of no-reflow. One study reported no cases of no- ischemic complications. Bivalirudin has been compared with
reflow in the verapamil treated population compared with 33.3% heparin in SVG PCI with adjunctive EPDs. The use of bivalir-
in the placebo group (39). Among patients pretreated with udin was safe and was associated with a trend toward fewer in-
intragraft nicardipine, only 4.4% had a CK-MB >3 times normal hospital non–Q wave MI, TVR, and vascular complications and
and the total MACE at 30 days was 4.4%, with no deaths, MI or a significantly lower rate of periprocedural CK-MB increases
repeat TVR from hospital discharge to 30 days (39). Prophylactic >5 times normal (44).
intragraft administration of a vasodilator can decrease the rates
of no-reflow and periprocedural elevations of CK-MB in patients Debulking Prior to SVG Stenting
treated without EPDs, however the role of medical pretreatment Mechanical approaches to plaque removal before stenting
in conjunction with EPDs has not been studied. Combined include directional atherectomy, excimer laser angioplasty, and
mechanical and medical therapy may offer an additive benefit extraction atherectomy (12,14,45). The CAVEAT II trial random-
for reducing MACE and long-term mortality after SVG PCI. ized patients with SVG lesions to treatment with directional
Clearly, adjunctive therapies in addition to EPD are needed to atherectomy vs. PTCA. Directional atherectomy was associated
further reduce the high adverse event rates after SVG PCI. with an increased incidence of distal embolization (13.4% vs.
5.1%) and non–Q wave MI (10.1% vs. 5.8%) (12). These disap-
Antithrombotic and Thrombolytic Therapy pointing results dampened any enthusiasm for the use of direc-
Given the consistent demonstration that GP IIb/IIIa receptor tional atherectomy in the treatment of SVG disease. The
antagonists improve procedural safety in native coronary transluminal extraction catheter combines both atherectomy
arteries, many operators routinely administer a GP IIb/IIIa and thrombectomy features and has been used in diffusely
receptor antagonist before SVG PCI. As discussed above, admin- diseased, thrombus-laden SVG, with the aim of minimizing
istration of these agents in nonrandomized trials has been distal embolization. Safian analyzed the results of extraction
[2/7/010/9:19:38] [469–481]
Figure 36.6 A post–coronary artery bypass surgery patient presented with AMI. Graft angiography demonstrated total occlusion of the SVG to
the left posterolateral artery. (A) Following passage of a guidewire through the SVG into the native coronary artery, angiography revealed a
degenerated SVG with large thrombus burden. (B) The AngioJetTM (Possis Medical, Inc., Minneapolis, Minnesota, U.S.) is positioned in the distal
SVG. (C) After being treated with the Angiojet, the large filling defects are no longer evident. (D) The distal SVG was stented with adjunctive distal
protection using the GuardwireTM (Medtronic, Inc., Santa Rosa, California, U.S.). Abbreviation: SVG, saphenous vein graft.
atherectomy in 146 patients with SVG disease and reported a occlusion (14,46). Ahmed compared debulking with excimer
21% incidence of acute angiographic complication including laser prior to stand-alone stenting for treatment of aorto-ostial
distal embolization (11.3%), no-reflow (4.4%), and abrupt closure SVG lesions and found no difference in procedural safety or
(5%) (19). Results from the X-SIZERTM (EV3, Inc., Plymouth, efficacy (18). The ATLAS study randomized patients with acute
Minnesota, U.S.) for Treatment of Thrombus and Atherosclerosis coronary syndromes undergoing SVG PCI to acolysis (ultra-
in Coronary Interventions Trial (X-TRACT) comparing short- sound thrombolysis) or abciximab. The acolysis probe delivers
term outcomes in patients with SVG disease randomized to a therapeutic level of ultrasound which leads to a cavitation
stenting alone or pretreatment with the X-SIZER device (Endi- effect resulting in a vortex that pulls thrombus toward the
COR Medical, Inc., San Clemente, California, U.S.) showed a catheter tip. The study was terminated prematurely because of
benefit of this atherothrombectomy device (46). Thirty-day a significantly higher number of adverse clinical outcomes in
events included one death (2.0%), Q- or non–Q wave MI in the acolysis arm, with a 25% 30-day incidence of MACE with
4.0%, TVR in 6.0%, and any MACE in 6.0%. Pretreatment with acolysis compared with 12% with abciximab (p ¼ 0.036) (47).
the X-Sizer was associated with a lower incidence of large MI and
need for bailout GP IIb/IIIa receptor blockade. Embolic Protection
Adjunctive excimer laser before balloon angioplasty of As the widespread use of stents have consistently improved
SVG has been evaluated in retrospective studies, however, its acute angiographic outcome, considerable attention has been
use has been plagued by a high rate of restenosis and total focused on reducing the high rate of no-reflow and
[2/7/010/9:19:38] [469–481]
death, MI, emergency CABG, or TLR at 30 days was reduced by

42% in patients assigned to the PercuSurge EPD (Fig. 36.9). This
reduction in adverse events was driven primarily by a lower
incidence of no-reflow and MI. These findings were indepen-
dent of GP IIb/IIIa inhibitor use. Use of distal EPDs was
associated with a trend toward reduced mortality (1.0% vs.
2.3%). Interestingly, in a subset analysis of the SAFER trial,
embolic protection was associated with significantly lower
rates of MACE, even in patients that were directly stented
(14.1% in the nonprotected group vs. 5.5% in the protected
group, p < 0.001). The SAFER trial served as an important
“proof of concept,” establishing EPD technology as the first
adjunctive therapy to dramatically improve the safety of SVG
stenting. Embolic protection with the Guardwire was also
found to be highly cost effective (49).
A newer generation occlusive device is the ProxisTM
(St. Jude Medical, St. Paul, Minnesota, U.S.) device which
occludes inflow proximal to the target lesion (Fig. 36.7). Inflation
of the balloon interrupts antegrade flow in the SVG during the
PCI. The stagnated column of blood can be aspirated throughout
the procedure to remove debris particles and must be done
before restoring flow. Advantages to the Proxis system include
choice of guidewires, the establishment of protection before a
wire or bulky device crosses the lesion, and the ability to protect
arteriess with multiple side branches or distal lesions that pro-
hibit the use other distal EPDs (50). The Proximal Protection
During Saphenous Vein Graft Intervention Using the Proxis
Embolic Protection System (PROXIMAL) trial was a multicenter,
prospective randomized trial comparing proximal protection
with distal protection with either a balloon occlusion or a filter
wire. The primary composite endpoint of death, MI, or TVR at 30
days by intention-to-treat analysis occurred in 10% of the distal
protection group and 9.2% of the proximal protection group (p
Figure 36.7 Mechanism of action of the embolic retrieval devices. for noninferiority ¼ 0.0061). Secondary analyses suggested that
Distal occlusion (top), distal filtration (middle), and proximal occlu- among patients whose lesions were amenable to either method of
sion (bottom). Source: From Ref. 55. protection, proximal protection was associated with a numeri-
cally lower, although not statistically significant, 30-day MACE
rate. This study established noninferiority of the Proxis device in
preventing 30-day MACE and suggested a possible benefit of
periprocedural MI commonly observed during SVG PCI. Once protecting the SVG prior to crossing the lesion with a wire or
significant distal embolization has occurred, the likelihood of device (51).
myonecrosis and its sequelae are high. Thus, considerable A second class of distal EPDs are embolic protection
research has focused on prevention of embolization of throm- filters that trap debris in semi-porous membranes while allow-
botic and atheromatous material by EPDs. These devices are ing antegrade perfusion (Fig. 36.7). Filter wires offer several
grouped into three classes: distal occlusion balloons, distal theoretic benefits over balloon occlusion devices. Since flow is
filters, and proximal occlusion balloons (Fig. 36.7). The proto- maintained during stent deployment, angiographic landmarks
typic balloon occlusion distal EPD is the GuardwireTM (Med- can guide accurate stent placement. In addition, maintenance of
tronic, Inc., Santa Rosa, California, U.S.). A nitinol 0.014-in flow reduces the occurrence of prolonged no-flow ischemia,
diameter hypotube inside a guidewire functions as an inflation which may result in hemodynamic compromise if the treated
lumen for a compliant balloon located near the distal end of the SVG subtends a large amount of myocardium or if left ven-
wire. After the balloon is inflated in the distal SVG, balloon tricular function is significantly compromised. Potential limi-
angioplasty or stenting is performed over the Guardwire, tations include flow compromise from filter sludging, failure to
which serves as the interventional guidewire throughout the block soluble mediators of no-reflow, embolization of particles
procedure (Fig. 36.8). A hollow-lumen catheter attached to a during filter recovery, and the embolization of small (<100 mm)
large syringe is passed over the wire, allowing aspiration of particles. The randomized FIRE trial compared the filter-based
material dislodged from the SVG lumen. Webb reported favor- FilterWire EXTM (Boston Scientific, Natick, Massachusetts, U.S.)
able outcomes in the first 23 procedures performed with to distal protection with the Guardwire (52). The 30-day out-
PercuSurge Guardwire, with an incidence of periprocedural comes of SVG PCI using the FilterWire EX system were non-
MI of only 3.7% (30). The pivotal Saphenous Vein Graft inferior to those with the Guardwire balloon occlusion system.
Angioplasty Free of Emboli Randomized (SAFER) trial The postprocedural measures of epicardial blood flow and
randomized 801 patients who underwent SVG stenting to con- angiographic complications were similar between the two
ventional therapy or distal protection using the PercuSurge groups, although there was a slightly higher rate of GP IIb/
Guardwire system (48). The primary composite endpoint of IIIa inhibitor use for bailout therapy in the GuardWire group.
[2/7/010/9:19:38] [469–481]
Figure 36.8 The PercuSurge embolic protection system. The GuardWireTM (Medtronic, Inc., Santa Rosa, California, U.S.) is positioned in
the distal SVG (left ). With the distal protection balloon inflated, a stent is deployed across the stenosis (middle). Embolic debris liberated
during balloon expansion is aspirated with the ExportTM (Medtronic, Inc., Santa Rosa, California, U.S.) catheter (right ). Abbreviation: SVG,
saphenous vein graft.
Figure 36.9 Survival free from MACE (at 30 days):

eventfree survival 1.5 SEM in all patients. Source:
From Ref. 48.
The six-month outcomes from the FIRE trial revealed a 3.5% all tion. Rogers et al. characterized the size and volume of debris
cause mortality in the entire study population, MI in 12% of retrieved from either a balloon occlusion device or a filter wire
patients at six months, and TVR in 9%. Although the outcomes during SVG stenting and demonstrated that with both methods
between the two groups were similar at six months, the rates of of protection, most particles were <100 mm in the longest
MACE in both groups were not insignificant (53). dimension. Additionally, both types of EPDs had a similar
The favorable results of the SAFER and FIRE trials affirm total embolic load per lesion supporting the findings of equiv-
the causal link between distal embolization and clinical events alency from FIRE trial, assuming that soluble mediators are not
following SVG stenting, and establish the practice standard for major contributors to MACE poststenting (54). Second genera-
prevention of these complications. However, the six-month tion EPDs have incorporated flush and extraction technology to
outcomes from the FIRE trial reinforce the fact that long-term balloon occlusion systems, and newer filter protection systems
outcomes after SVG PCI still remain poor. The FIRE trial such as the SpiderTM (EV3, Inc., Plymouth, Minnesota, U.S.)
showed parity between balloon occlusion and filter wires but allow the operator to cross the stenosis with any 0.014-in
did not analyze the particulate matter retrieved after interven- guidewire and deliver the filter system over that wire. Despite
[2/7/010/9:19:38] [469–481]
these technologic improvements, major adverse events still which randomized patients to either the SymbiotTM (Boston
occur in 5% to 10% of patients, and half of SVG interventions Scientific, Natick, Massachusetts, U.S.) covered stent or BMS
are performed without EPDs (55). showed no benefit of the covered stents in the treatment of
degenerated SVG. The rates of binary restenosis in the stented
segment were similar (29.1% Symbiot vs. 21.9% BMS, p ¼ 0.17),
Covered Stents however more patients in the Symbiot group had binary
Another approach to the prevention of distal embolization
restenosis at the proximal edge (9.0% Symbiot vs. 1.8% BMS,
following SVG stenting is the use of covered stents. Stents
p ¼ 0.0211). There was no difference in MACE between groups
covered with a variety of organic and synthetic compounds
(30.6% Symbiot vs. 26.6% BMS, p ¼ 0.43) (59). The hypothesis
have been deployed to exclude aneurysms and seal perfora-
that covered stents may reduce periprocedural complications
tions. The GraftMasterTM (Abbott Vascular, Abbott Park, Illi-
by potentially preventing distal embolization and serve as a
nois, U.S.) polytetrafluoroethylene (PTFE)-covered stent was
possible barrier to cell migration, thereby reducing restenosis,
approved by the U.S. Food and Drug Administration under a
appears to have been invalidated.
Humanitarian Device Exemption for treatment of coronary
perforation (Fig. 36.10). It was anticipated that covered stents
might also exclude friable atheroma thereby reducing athe-
CLINICAL OUTCOMES AND SPECIAL
roembolic complications and no-reflow states during SVG
CONSIDERATIONS
stenting. In a multicenter series of 109 consecutive patients,
Restenosis, Recurrent Events, and Long-Term
Baldus reported successful deployment of the JoMed PTFE
Outcome Following SVG Stenting
stent in all but one of 109 patients with an incidence of
Although the SAVED trial demonstrated improved the long-
periprocedural MI of only 1% (56). In a small series of patients
term outcome of stenting compared with balloon angioplasty,
treated with either a PTFE-covered stent or a BMS, those
and newer technologies such as EPDs have improved procedural
treated with a PTFE-covered stent had a lower incidence of
safety, the long-term outcomes after SVG PCI are still subopti-
non–Q wave MI (57). However, tempering this initial enthusi-
mal. Recurrent ischemic events are quite common (>50% by
asm for the use of PTFE-covered stents in SVG was a report
18 months) in patients who have undergone successful SVG
from the Randomized Evaluation of Covered Stent in Saphe-
stenting (16,60–63). There is controversy as to whether the inci-
nous Vein Graft (RECOVERS) trial in which 201 patients with
dence, time course, and biology of restenosis differs in SVG
SVG disease were randomized to treatment with either the
compared with native coronary arteries following stenting. More
Jostent Flex BMS or the PTFE-covered JoMed stent. The inci-
clinical events occur beyond 6 to 12 months in stented SVG
dence of no-reflow (5.8% vs. 2.1%) and 30-day major adverse
compared with native coronary arteries, suggesting that the
events (15.3% vs. 12.4%) were higher in patients randomized to
restenotic hazard in SVG has a prolonged time course. Whereas
the PTFE-covered stent (58). A more recent trial, SYMBIOT,
the incidence of TVR and non-TVR is higher in patients with
SVG, the rate of revascularization driven by failure of the target
lesion is similar to that of native coronary arteries (61). Thus, the
incidence of clinical events approaches 50% by five years because
of the progression of disease at nontarget sites within the treated
SVG, the attrition of other SVG, and the progression of native
coronary artery disease (60). Depre and colleagues examined the
histology of previously stented SVG and reported “long-term
restenosis” occurs in at least 30% of cases and results from a
combination of cellular hyperplasia, thrombosis, and progressive
atherosclerosis (64). Another histologic study found abnormal
adherence of inflammatory cells and platelets as late as 10 months
after implantation suggesting a propensity toward late throm-
bosis (65). While this may partly explain the observation that late
reocclusion often accompanies in-stent restenosis, it is important
to remember that other factors such as the absence of side
branches and the rapid progression of atherosclerosis at other
loci within the SVG may also predispose to thrombotic occlusion.
The presence of diabetes mellitus, nonobstructive stenoses in
other SVG segments, restenosis, placement of multiple stents,
SVG degeneration, peripheral vascular disease, and congestive
heart failure were all predictors of MAE following SVG stenting
(62,66–68). Although female gender was associated with peri-
procedural and 30-day MACE, clinical outcome at 12 months
was independent of gender (69).
Predictors of angiographic restenosis following SVG stent-
ing include previously dilated lesions, smaller reference diame-
ter, diabetes mellitus, and smaller posttreatment diameter
stenosis (17). SVG in-stent restenosis has been treated by a
Figure 36.10 The GraftMasterTM polytetrafluoroethylene-covered variety of approaches including repeat balloon dilation, excimer
stent. laser angioplasty; and directional, rotational, and extraction
atherectomy (62,70,71). Repeat PCI does not appear to entail
[2/7/010/9:19:38] [469–481]
the high risk of distal embolization associated with treatment of from the SAFER trial identified angiographic thrombus as an
de novo lesions. In a series of 54 patients who underwent important predictor of MACE. While the use of GP IIb/IIIa
treatment of in-stent restenosis, distal embolization resulting in receptor antagonists has not been associated with the same
MACE was not observed (72). Despite excellent acute angio- benefit during SVG stenting compared with treatment of native
graphic and clinical outcome, the incidence of recurrent reste- coronary arteries, their use should be considered if a large
nosis approaches 50%. thrombus burden is angiographically apparent, or suspected
clinically, for example, when intervention is performed for an
acute coronary syndrome, or the graft is freshly occluded.
Contemporary Approaches to the Technique Intragraft administration of the abciximab may reduce throm-
of SVG Stenting bus burden. Alternatively, the use of a mechanical thrombec-
First generation coronary devices such as the Gianturco-Roubin tomy device such as the Excisor or a thrombus aspiration
and Palmaz-Schatz stents were designed to treat focal lesions in catheter such as the Export catheter with an EPD may be
native coronary arteries. SVG are larger and contain more helpful.
diffuse, friable atheroma than native coronary arteries (8). Generally, with the propensity of SVG atheroma to
Consequently, multiple stents were commonly required and embolize, a “less is better” approach to stenting may be
plaque prolapse was frequently observed when stents designed warranted even when the EPD is employed. Primary stenting
to treat vessels 3 to 4 mm in diameter were overexpanded to (i.e., without predilatation) should be strongly considered.
treat larger SVG. In an effort to provide better scaffolding by Webb has shown that primary stenting is associated with a
increasing the metal to surface area ratio, hand-crimped larger reduced volume of embolized debris (30). Contemporary
biliary stents were used to treat diseased SVG (73). However, stents are mounted on balloons that can be inflated to high
these stents were difficult to deliver because of high profile, pressure. Thus, if the operator chooses the correct stent size,
rigidity, and suboptimal anchoring of the stent to the delivery postdilatation can be avoided in most cases. However, occa-
balloon. This necessitated the use of 8-Fr or 9-Fr guiding sionally one encounters a noncompliant lesion resulting from
catheters, lesion predilatation, and relatively stiff guidewires significant adventitial fibrosis that requires high-pressure
for delivery. This prompted several vendors to develop modi- postdilation with a noncompliant balloon. It is important
fied balloon-expandable stents to treat larger vessels. These that all subsequent inflations be performed with an EPD as
balloon-expandable delivery systems are the platforms for both embolism and slow flow can occur even with postdilation.
BMS and DES. These large vessel stents are compatible with Occasionally, balloon rupture or severe barotrauma may lead
6-Fr guiding catheters, have relatively low crimped profiles and to perforation of the vein graft. Usually, small perforations do
greater flexibility, and increased surface area coverage to min- not lead to hemodynamic compromise as they rarely cause
imize plaque prolapse. Additionally, with the development of pericardial tamponade and extensive fibrosis often seals the
DES and the decreased rates of restenosis seen in native coro- perforation. When the perforation is large, or associated with
nary arteries, the use of DES in SVG PCI has rapidly increased. hemodynamic instability, the perforation can be reliably
Guiding catheter selection is an important component of sealed with a PTFE-covered stent (74). Before the advent of
successful SVG stenting. Ideally, the operator should choose a EPDs, moderate stenoses were often left untreated to avoid
guiding catheter whose shape allows selective, coaxial intuba- additional plaque manipulation. However, since these non-
tion of the SVG. For SVG anastomosed to branches of the left flow-limiting plaques may be associated with rapid disease
coronary artery, a Right Judkins catheter will generally engage progression and future ischemic events (66,68), the availability
the SVG and is usually adequate for diagnostic angiography. of reliable, easy to use EPDs allows the operator to take a more
However, the Right Judkins guiding catheter often provides proactive, aggressive approach to these lesions. When patients
minimal backup support and frequently disengages from the with SVG stents develop recurrent symptoms, angiography
ostium during balloon or stent delivery. A Hockey Stick or should be performed since progressive flow impairment may
Amplatz Left guiding catheter will allow deeper intubation of result in thrombotic occlusion of the SVG. Unless clinically
the SVG especially if the SVG has a vertical take-off or if contraindicated because of hypotension, we routinely pretreat
maximum support is needed to facilitate delivery of longer with vasodilators to decrease rates of no-reflow. One series
stents or through tortuous segments. Care must be taken when has reported a low rate of adverse events with vasodilator
manipulating this aggressive catheter as the post-CABG patient pretreatment (39).
often has friable atheroma in the ascending aorta that can
dislodge when traumatized. Occasionally, when grafting the Treatment of Early (<1 Month) SVG Failure
left circumflex artery, the surgeon will place the aortic anasto- Early ischemic events may manifest as recurrent chest pain, MI,
mosis on the posterior surface of the aorta. These SVG can be hemodynamic instability, or heart failure and are usually
successfully cannulated with either a multipurpose or Right precipitated by SVG thrombosis resulting from a combination
Judkins guiding catheter. SVG to the right coronary artery of a prothrombotic milieu and technical misadventure. Treat-
typically are sutured to the posterior surface of the aorta and ment of thrombosed SVG usually involves a combination of
usually have a downward orientation. The multipurpose or platelet inhibition with GP IIb/IIIa receptor antagonists and a
AR-1 guiding catheters generally provide adequate visualiza- thienopyridine, and thrombectomy, followed by balloon dila-
tion and support. tation and or stent placement, usually at the aorto-ostial or
Given the dramatic reduction in no-reflow and its clinical distal anastomosis (75). Since the flow-limiting lesion is usually
sequelae observed in both the SAFER and FIRE trials, the use of located at a freshly sutured site, excessive pressure needed to
an EPD is recommended. Exceptions to this recommendation expand the stent may disrupt the sutures resulting in severe
include SVG less than a year old or treatment of in-stent hemorrhage or tamponade. This risk may be highest for stents
restenosis. Stenting of SVG with large amounts of angiographic placed at the distal anastomosis when sutures are placed
thrombus can be especially challenging. Subgroup analysis through the back wall of the native coronary artery.
[2/7/010/9:19:38] [469–481]
CONCLUSIONS 14. Strauss BH, Natarajan MK, Batchelor WB, et al. Early and late
SVG disease after CABG is common. One year after surgery, up quantitative angiographic results of vein graft lesions treated by
to 15% of SVG are occluded and only 61% of SVG are patent at excimer laser with adjunctive balloon angioplasty. Circulation
10 years. Among the patent SVG, many are diffusely diseased 1995; 92:348–356.
15. Fenton SH, Fischman DL, Savage MP, et al. Long-term angio-
and repeat CABG is associated with increased morbidity and
graphic and clinical outcome after implantation of balloon-
mortality. Therefore, PCI is the preferred revascularization expandable stents in aortocoronary saphenous vein grafts. Am J
strategy for SVG disease. However, it is limited by a substantial Cardiol 1994; 74:1187–1191.
risk of MACE caused mainly by periprocedural MI as a result of 16. Piana RN, Moscucci M, Cohen DJ, et al. Palmaz-Schatz stenting for
no-reflow and distal embolization. Treatment of SVG disease is treatment of focal vein graft stenosis: immediate results and long-
associated with a high rate of restenosis ranging from 20% to term outcome. J Am Coll Cardiol 1994; 23:1296–1304.
37% and high rates of TVR. Recent advances in SVG PCI have 17. Wong SC, Baim DS, Schatz RA, et al. Immediate results and late
included the development of EPDs that have reduced the rates outcomes after stent implantation in saphenous vein graft lesions:
of 30-day MACE by 42%. Nonetheless, SVG PCI is still associ- the multicenter U.S. Palmaz-Schatz stent experience. The Palmaz-
ated with worse long-term outcomes than native coronary Schatz Stent Study Group. J Am Coll Cardiol 1995; 26:704–712.
18. Ahmed JM, Hong MK, Mehran R, et al. Comparison of debulking
artery PCI and research is needed to develop pharmacologic
followed by stenting versus stenting alone for saphenous vein
as well as mechanical devices to improve the outcomes of SVG graft aortoostial lesions: immediate and one-year clinical out-
PCI. comes. J Am Coll Cardiol 2000; 35:1560–1568.
19. Savage MP, Douglas JS Jr., Fischman DL, et al. Stent placement
compared with balloon angioplasty for obstructed coronary
REFERENCES bypass grafts. Saphenous Vein De Novo Trial Investigators. N
1. Favaloro RG. Saphenous vein autograft replacement of severe Engl J Med 1997; 337:740–747.
segmental coronary artery occlusion: operative technique. Ann 20. Gruberg L, Hong MK, Mehran R, et al. In-hospital and long-term
Thorac Surg 1968; 5:334–339. results of stent deployment compared with balloon angioplasty
2. Shelton ME, Forman MB, Virmani R, et al. A comparison of for treatment of narrowing at the saphenous vein graft distal
morphologic and angiographic findings in long-term internal anastomosis site. Am J Cardiol 1999; 84:1381–1384.
mammary artery and saphenous vein bypass grafts. J Am Coll 21. Hanekamp CE, Koolen JJ, Den Heijer P, et al. Randomized study
Cardiol 1988; 11:297–307. to compare balloon angioplasty and elective stent implantation in
3. Cameron A, Davis KB, Green GE, et al. Clinical implications of venous bypass grafts: the Venestent study. Catheter Cardiovasc
internal mammary artery bypass grafts: the Coronary Artery Interv 2003; 60:452–457.
Surgery Study experience. Circulation 1988; 77:815–819. 22. Vermeersch P, Agostoni P, Verheye S, et al. Randomized double-
4. Chen L, Theroux P, Lesperance J, et al. Angiographic features of blind comparison of sirolimus-eluting stent versus bare-metal
vein grafts versus ungrafted coronary arteries in patients with stent implantation in diseased saphenous vein grafts: six-month
unstable angina and previous bypass surgery. J Am Coll Cardiol angiographic, intravascular ultrasound, and clinical follow-up of
1996; 28:1493–1499. the RRISC Trial. J Am Coll Cardiol 2006; 48:2423–2431.
5. Morrison DA, Sethi G, Sacks J, et al. Percutaneous coronary 23. Lee MS, Shah AP, Aragon J, et al. Drug-eluting stenting is superior
intervention versus repeat bypass surgery for patients with med- to bare metal stenting in saphenous vein grafts. Catheter Cardio-
ically refractory myocardial ischemia: AWESOME randomized vasc Interv 2005; 66:507–511.
trial and registry experience with post-CABG patients. J Am 24. Ge L, Iakovou I, Sangiorgi GM, et al. Treatment of saphenous vein
Coll Cardiol 2002; 40:1951–1954. graft lesions with drug-eluting stents: immediate and midterm
6. Goldman S, Zadina K, Moritz T, et al. Long-term patency of outcome. J Am Coll Cardiol 2005; 45:989–994.
saphenous vein and left internal mammary artery grafts after 25. Vignali L, Saia F, Manari A, et al. Long-term outcomes with drug-
coronary artery bypass surgery: results from a Department of eluting stents versus bare metal stents in the treatment of saphenous
Veterans Affairs Cooperative Study. J Am Coll Cardiol 2004; vein graft disease (results from the REgistro Regionale AngiopLas-
44:2149–2156. tiche Emilia-Romagna registry). Am J Cardiol 2008; 101:947–952.
7. Canos DA, Mintz GS, Berzingi CO, et al. Clinical, angiographic, 26. Pucelikova T, Mehran R, Kirtane AJ, et al. Short- and long-term
and intravascular ultrasound characteristics of early saphenous outcomes after stent-assisted percutaneous treatment of saphenous
vein graft failure. J Am Coll Cardiol 2004; 44:53–56. vein grafts in the drug-eluting stent era. Am J Cardiol 2008; 101:63–68.
8. de Feyter PJ, van Suylen RJ, de Jaegere PP, et al. Balloon angio- 27. Vermeersch P, Agostoni P, Verheye S, et al. Increased late mor-
plasty for the treatment of lesions in saphenous vein bypass grafts. tality after sirolimus-eluting stents versus bare-metal stents in
J Am Coll Cardiol 1993; 21:1539–1549. diseased saphenous vein grafts: results from the randomized
9. Platko WP, Hollman J, Whitlow PL, et al. Percutaneous trans- DELAYED RRISC Trial. J Am Coll Cardiol 2007; 50:261–267.
luminal angioplasty of saphenous vein graft stenosis: long-term 28. Hong MK, Mehran R, Dangas G, et al. Creatine kinase-MB enzyme
follow-up. J Am Coll Cardiol 1989; 14:1645–1650. elevation following successful saphenous vein graft intervention
10. Reeves F, Bonan R, Cote G, et al. Long-term angiographic follow- is associated with late mortality. Circulation 1999; 100:2400–2405.
up after angioplasty of venous coronary bypass grafts. Am Heart J 29. Carrozza JP Jr., Schatz RA, George CJ, et al. Acute and long-term
1991; 122:620–627. outcome after Palmaz-Schatz stenting: analysis from the New
11. Plokker HW, Meester BH, Serruys PW. The Dutch experience in Approaches to Coronary Intervention (NACI) registry. Am J
percutaneous transluminal angioplasty of narrowed saphenous Cardiol 1997; 80:78K–88K.
veins used for aortocoronary arterial bypass. Am J Cardiol 1991; 30. Webb JG, Carere RG, Virmani R, et al. Retrieval and analysis of
67:361–366. particulate debris after saphenous vein graft intervention. J Am
12. Holmes DR Jr., Topol EJ, Califf RM, et al. A multicenter, random- Coll Cardiol 1999; 34:468–475.
ized trial of coronary angioplasty versus directional atherectomy 31. Motwani JG, Topol EJ. Aortocoronary saphenous vein graft dis-
for patients with saphenous vein bypass graft lesions. CAVEAT-II ease: pathogenesis, predisposition, and prevention. Circulation
Investigators. Circulation 1995; 91:1966–1974. 1998; 97:916–931.
13. Safian RD, Grines CL, May MA, et al. Clinical and angiographic 32. Golino P, Piscione F, Benedict CR, et al. Local effect of serotonin
results of transluminal extraction coronary atherectomy in saphe- released during coronary angioplasty. N Engl J Med 1994;
nous vein bypass grafts. Circulation 1994; 89:302–312. 330:523–528.
[2/7/010/9:19:38] [469–481]
33. Sdringola S, Assali AR, Ghani M, et al. Risk assessment of slow or occlusion and aspiration system during percutaneous intervention
no-reflow phenomenon in aortocoronary vein graft percutaneous of diseased saphenous vein aorto-coronary bypass grafts. Circu-
intervention. Catheter Cardiovasc Interv 2001; 54:318–324. lation 2003; 108:548–553.
34. Piana RN, Paik GY, Moscucci M, et al. Incidence and treatment of 53. Halkin A, Masud AZ, Rogers C, et al. Six-month outcomes after
‘no-reflow’ after percutaneous coronary intervention. Circulation percutaneous intervention for lesions in aortocoronary saphenous
1994; 89:2514–2518. vein grafts using distal protection devices: results from the FIRE
35. Coolong A, Baim DS, Kuntz RE, et al. Saphenous vein graft stenting trial. Am Heart J 2006; 151:915.e1–915.e7.
and major adverse cardiac events: a predictive model derived from 54. Rogers C, Huynh R, Seifert PA, et al. Embolic protection with
a pooled analysis of 3958 patients. Circulation 2008; 117:790–797. filtering or occlusion balloons during saphenous vein graft stent-
36. Kirtane AJ HE, Metzger C, Breall JA, et al. Correlates of adverse ing retrieves identical volumes and sizes of particulate debris.
events during saphenous vein graft intervention with distal Circulation 2004; 109:1735–1740.
embolic protection: a PRIDE substudy. JACC Cardiovasc Interv 55. Mauri L, Rogers C, Baim DS. Devices for distal protection during
2008; 1:186–191. percutaneous coronary revascularization. Circulation 2006;
37. Kaplan BM, Benzuly KH, Kinn JW, et al. Treatment of no-reflow in 113:2651–2656.
degenerated saphenous vein graft interventions: comparison of 56. Baldus S, Koster R, Elsner M, et al. Treatment of aortocoronary
intracoronary verapamil and nitroglycerin. Cathet Cardiovasc vein graft lesions with membrane-covered stents: a multicenter
Diagn 1996; 39:113–118. surveillance trial. Circulation 2000; 102:2024–2027.
38. Weyrens FJ, Mooney J, Lesser J, et al. Intracoronary diltiazem for 57. Briguori C, De Gregorio J, Nishida T, et al. Polytetrafluoroethy-
microvascular spasm after interventional therapy. Am J Cardiol lene-covered stent for the treatment of narrowings in aorticocoro-
1995; 75:849–850. nary saphenous vein grafts. Am J Cardiol 2000; 86:343–346.
39. Fischell TA, Subraya RG, Ashraf K, et al. “Pharmacologic” distal 58. Stankovic G, Colombo A, Presbitero P, et al. Randomized evalu-
protection using prophylactic, intragraft nicardipine to prevent no- ation of polytetrafluoroethylene-covered stent in saphenous vein
reflow and non-Q-wave myocardial infarction during elective grafts: the Randomized Evaluation of polytetrafluoroethylene
saphenous vein graft intervention. J Invasive Cardiol 2007; 19:58–62. COVERed stent in Saphenous vein grafts (RECOVERS) Trial.
40. Roffi M, Mukherjee D, Chew DP, et al. Lack of benefit from Circulation 2003; 108:37–42.
intravenous platelet glycoprotein IIb/IIIa receptor inhibition as 59. Turco MA, Buchbinder M, Popma JJ, et al. Pivotal, randomized U.
adjunctive treatment for percutaneous interventions of aortocoro- S. study of the Symbiottrade mark covered stent system in
nary bypass grafts: a pooled analysis of five randomized clinical patients with saphenous vein graft disease: eight-month angio-
trials. Circulation 2002; 106:3063–3067. graphic and clinical results from the Symbiot III trial. Catheter
41. Barsness GW, Buller C, Ohman EM, et al. Reduced thrombus Cardiovasc Interv 2006; 68:379–388.
burden with abciximab delivered locally before percutaneous inter- 60. Frimerman A, Rechavia E, Eigler N, et al. Long-term follow-up of
vention in saphenous vein grafts. Am Heart J 2000; 139:824–829. a high risk cohort after stent implantation in saphenous vein
42. Teirstein PS, Mann JT III, Cundey PE Jr., et al. Low- versus high- grafts. J Am Coll Cardiol 1997; 30:1277–1283.
dose recombinant urokinase for the treatment of chronic saphe- 61. Laham RJ, Carrozza JP, Berger C, et al. Long-term (4- to 6-year)
nous vein graft occlusion. Am J Cardiol 1999; 83:1623–1628. outcome of Palmaz-Schatz stenting: paucity of late clinical stent-
43. Kuntz RE, Baim DS, Cohen DJ, et al. A trial comparing rheolytic related problems. J Am Coll Cardiol 1996; 28:820–826.
thrombectomy with intracoronary urokinase for coronary and 62. Le May MR, Labinaz M, Marquis JF, et al. Predictors of long-term
vein graft thrombus [the Vein Graft AngioJet Study (VeGAS 2)]. outcome after stent implantation in a saphenous vein graft. Am J
Am J Cardiol 2002; 89:326–330. Cardiol 1999; 83:681–686.
44. Rha SW, Kuchulakanti PK, Pakala R, et al. Bivalirudin versus 63. de Jaegere PP, van Domburg RT, Feyter PJ, et al. Long-term
heparin as an antithrombotic agent in patients who undergo clinical outcome after stent implantation in saphenous vein grafts.
percutaneous saphenous vein graft intervention with a distal J Am Coll Cardiol 1996; 28:89–96.
protection device. Am J Cardiol 2005; 96:67–70. 64. Depre C, Havaux X, Wijns W. Pathology of restenosis in saphe-
45. Braden GA, Xenopoulos NP, Young T, et al. Transluminal extrac- nous bypass grafts after long-term stent implantation. Am J Clin
tion catheter atherectomy followed by immediate stenting in treat- Pathol 1998; 110:378–384.
ment of saphenous vein grafts. J Am Coll Cardiol 1997; 30:657–663. 65. van Beusekom HM, van der Giessen WJ, van Suylen R, et al.
46. Bittl JA, Sanborn TA, Yardley DE, et al. Predictors of outcome of Histology after stenting of human saphenous vein bypass grafts:
percutaneous excimer laser coronary angioplasty of saphenous observations from surgically excised grafts 3 to 320 days after
vein bypass graft lesions. The Percutaneous Excimer Laser Coro- stent implantation. J Am Coll Cardiol 1993; 21:45–54.
nary Angioplasty Registry. Am J Cardiol 1994; 74:144–148. 66. Keeley EC, Velez CA, O’Neill WW, et al. Long-term clinical
47. Singh M, Rosenschein U, Ho KK, et al. Treatment of saphenous outcome and predictors of major adverse cardiac events after
vein bypass grafts with ultrasound thrombolysis: a randomized percutaneous interventions on saphenous vein grafts. J Am Coll
study (ATLAS). Circulation 2003; 107:2331–2336. Cardiol 2001; 38:659–665.
48. Baim DS, Wahr D, George B, et al. Randomized trial of a distal 67. Ahmed JM, Hong MK, Mehran R, et al. Influence of diabetes
embolic protection device during percutaneous intervention of mellitus on early and late clinical outcomes in saphenous vein
saphenous vein aorto-coronary bypass grafts. Circulation 2002; graft stenting. J Am Coll Cardiol 2000; 36:1186–1193.
105:1285–1290. 68. Ellis SG, Brener SJ, DeLuca S, et al. Late myocardial ischemic
49. Cohen DJ, Murphy SA, Baim DS, et al. Cost-effectiveness of distal events after saphenous vein graft intervention–importance of ini-
embolic protection for patients undergoing percutaneous inter- tially “nonsignificant” vein graft lesions. Am J Cardiol 1997;
vention of saphenous vein bypass grafts: results from the SAFER 79:1460–1464.
trial. J Am Coll Cardiol 2004; 44:1801–1808. 69. Ahmed JM, Dangas G, Lansky AJ, et al. Influence of gender on
50. Cragun DT, Heuser RR. Embolic protection devices in saphenous early and one-year clinical outcomes after saphenous vein graft
vein graft interventions. J Interv Cardiol 2006; 19:525–529. stenting. Am J Cardiol 2001; 87:401–405.
51. Mauri L, Cox D, Hermiller J, et al. The PROXIMAL trial: proximal 70. Virk SJ, Bellamy CM, Perry RA. Transluminal extraction atherec-
protection during saphenous vein graft intervention using the tomy for stent restenosis in a saphenous vein bypass graft. Eur
Proxis Embolic Protection System: a randomized, prospective, Heart J 1997; 18:350–351.
multicenter clinical trial. J Am Coll Cardiol 2007; 50:1442–1449. 71. Dangas G, Mehran R, Lansky AJ, et al. Acute and long-term
52. Stone GW, Rogers C, Hermiller J, et al. Randomized comparison results of treatment of diffuse in-stent restenosis in aortocoronary
of distal protection with a filter-based catheter and a balloon saphenous vein grafts. Am J Cardiol 2000; 86:777–779, A6.
[2/7/010/9:19:38] [469–481]
72. Ashby DT, Dangas G, Aymong EA, et al. Effect of percutaneous 74. Hernandez-Antolin RA, Banuelos C, Alfonso F, et al. Successful
coronary interventions for in-stent restenosis in degenerated sealing of an angioplasty-related saphenous vein graft rupture
saphenous vein grafts without distal embolic protection. J Am with a PTFE-covered stent. J Invasive Cardiol 2000; 12:589–593.
Coll Cardiol 2003; 41:749–752. 75. Hanratty CG, Koyama Y, Ward MR. Angioplasty and stenting of
73. Friedrich SP, Davis SF, Kuntz RE, et al. Investigational use of the the distal coronary anastomosis for graft failure immediately after
Palmaz-Schatz biliary stent in large saphenous vein grafts. Am J coronary artery bypass grafting. Am J Cardiol 2002; 90:1009–1011.
Cardiol 1993; 71:439–441.
[2/7/010/9:21:13] [482–491]
37
Emboli protection devices, atherectomy,

thrombus aspiration devices
Fernando Cura
INTRODUCTION or inherent poor vascular ability to produce outward remodel-

A large armamentarium of medical devices is available for the ing and significant arterial luminal stenosis ensues. Despite the
percutaneous treatment of a variety of coronary atherosclerosis previous belief that the presence of high shear stress and
lesions. This chapter describes several aspects related to the continuous outward remodeling were only protective factors,
biological components of the atherosclerotic plaque and the their presence in conjunction with high plaque burden and a
rationale, clinical evidence, and indications for different inter- thin fibrous cap represent the classic scenario leading to a
ventional devices. The scope of this chapter will range from ruptured and/or eroded plaque (11). This metabolically active,
devices that prepare very hard atherosclerotic plaques prior to unstable plaque expresses several prothrombotic cytokines
stenting, for example, rotational atherectomy, to devices that leading to thrombus formation and vasoconstriction. This is
are either used to remove very soft, friable, or thrombotic the most common underlying coronary lesion among patients
lesions, such as thrombectomy, or those preventing distal with acute coronary syndromes. In particular, patients with
embolization during the intervention by temporary vessel ST-segment elevation acute myocardial infarction (STEMI)
occlusion or filter placement. have highly thrombotic lesions. In addition, these patients
have different levels of myocardial damage with microvascular
embolization, interstitial edema, and vasoconstriction (12).
PLAQUE COMPOSITION AND SELECTION On the other hand, coronary stenosis due primarily to
OF INTERVENTIONAL STRATEGY negative remodeling is associated with low plaque burden and
Normal human coronary arteries have three thin layers, thick fibrous cap, usually showing as stable coronary syn-
namely, the intima, media, and adventitia. The intima inter- dromes (13). Some of these patients have dense fibrocalcified
faces with the arterial lumen and is composed of endothelial coronary plaques, sometimes difficult to adequately address
cells. The media is composed of smooth muscle cells, collagen, with balloon angioplasty or stenting alone. This type of lesion
fibroblast, and intercellular matrix molecules, and regulates has been the target of numerous devices chiefly designed to
vascular contraction. The media is separated from the intima remove or alter plaque as an adjunctive measure prior to
and adventitia by the internal and external elastic membranes, stenting. Devices such as directional coronary atherectomy,
respectively. Coronary atherosclerosis is a complex inflamma- rotational atherectomy, excimer laser, and cutting balloon,
tory-infiltrative disorder affecting the three vessel layers (1). among others, have been tested (14,15). The two most com-
During life, the presence of acquired and inherent coronary monly used devices in current practice for severely fibrocalci-
artery disease risk factors triggers a series of events that leads to fied stable coronary lesions are rotational atherectomy and
atherosclerosis. Prior results from animal studies have sug- cutting balloon angioplasty.
gested that the first step in atherosclerosis is endothelial cell
dysfunction and lipoprotein particle accumulation at the intima
level (2). At this stage, lipid oxidation, local release of inter- Atherectomy Devices
leukins, cytokines, reactive oxygen species, and pro-oxidative Rotational atherectomy (16) is commercially available as the
enzymes generate white blood cell chemotaxis and adhesion to Rotablator System (Boston Scientific, Scimed, Maple Grove,
the endothelial surface (3–5). Leukocytes and monocytes pen- Minnesota, U.S.; Fig. 37.1) and consists of a nickel-plated, dia-
etrate endothelial cells into the arterial wall engulfing lipid mond chip-coated brass burr rotating at 140,000 to 190,000 rpm.
particles constituting foam cells (6,7). Then foam cells progress Eight burr sizes are available ranging from 1.25 to 2.5 mm.
to fatty streaks and ultimately to raised atheromatous plaques Rotational atherectomy works by the physical principle of
(1). Several other factors are also involved in the development differential cutting (17). This principle enables destruction of
of atherosclerosis. Laminar flow elicits atheroprotective proper- inelastic material such as atherosclerotic, calcified, and fibrotic
ties (8), while turbulent flow, especially at branch points, leads plaques while preventing normal elastic tissue from becoming
to atheromatous plaque formation. In fact, during plaque damaged. Debris dislodged after rotational atherectomy are less
growth, lumen reduction increases blood flow velocity, thus, than 12 mm, and should not create a significant impact in the
elevating shear stress. A large amount of evidence indicates epicardial and myocardial blood flow with careful technique.
that high levels of shear stress stimulate outward vascular However, CK-MB release (myocardial infarction) rates post
remodeling of the arterial wall, probably in an attempt to rotational atherectomy are significantly higher than after con-
preserve endoluminal dimensions (9,10). At a still unpredict- ventional angioplasty (18–20). Selection of burr size should not
able point in the atherosclerosis process, outward remodeling exceed a burr/artery diameter ratio of 0.7. In fact, for undilat-
may fail or just simply stop due to an excess of plaque burden able lesions, the use of very small burrs is less aggressive and
[2/7/010/9:21:13] [482–491]
ATHERECTOMY, EMBOLI PROTECTION, THROMBUS ASPIRATION DEVICES 483
Figure 37.1 The Rotablator System (A) burr, (B) advancer, and (C) control console (Boston Scientific, Scimed, Maple Grove, Minnesota, U.S.).
usually sufficient to physically alter the plaque surface, allow- risk of vessel trauma during device delivery to the lesion or
ing balloon dilatation and stenting. During the bare-metal stent balloon retrieval from the vessel. Usually, lower balloon infla-
era, substantial evidence indicated that despite achieving tion pressures (4–8 atm) are required. Currently, cutting balloon
greater acute gain with the use of rotational atherectomy, use is limited to undilatatable coronary artery stenosis, prefer-
restenosis remained unchanged and increased the risk of the ably with a reference vessel diameter ranging of 2.0 to 4.0 mm in
procedure due to distal embolization or slow flow (20). diameter, without extreme vessel tortuosity (32). Highly tortu-
Increased operator experience using shorter runs (<20 seconds) ous and angulated lesions are technically difficult to access due
with <5000 RPM decrements, the use of smaller burr size and to the presence of the longitudinal atherotomes in the cutting
device improvements have reduced periprocedural complica- balloon. Shorter cutting balloons (10 mm) are usually more
tions. Prophylactic use of intracoronary “Rota Flush Cocktails” deliverable than the longer (15 mm) ones in passing a moder-
with different concentrations of nitroglycerine, verapamil, and ately tortuous vessel. Cutting balloons can also be used in aorto-
heparin has improved the epicardial flow after rotational ostial lesions (33). Its anecdotal popularity for patients with
atherectomy (21). However, ventricular arrhythmias, coronary in-stent restenosis has been replaced with overwhelming data
dissection, and vessel perforation are still a concern, especially supporting the use of DES (34,35).
among patients with severe vessel tortuosity or angulated
lesions (22). Thus, the popularity of rotational atherectomy
has decreased in the past decade, and it is being predominantly Thrombus Removal and Prevention
utilized for undilatable lesions or delivery failures. Unfortu- of Distal Embolization
nately, the complexity of this device requires proper training Unstable coronary syndromes are associated with soft, friable,
and experience for the operator to be comfortable using this thrombotic and metabolically active atherosclerotic lesions (36).
unique tool. At sites of plaque formation there is substantial inflammation
Although intravascular ultrasound studies have demon- (37,38) with macrophages releasing proteolytic enzymes such
strated that suboptimal deployment, incomplete apposition, as matrix metalloproteinases responsible for thinning fibrous
and failure to achieve adequate stent dimensions may lead to cap leading to plaque erosion or rupture and thrombus forma-
stent thrombosis (23–25) and repeat target vessel revascularization (39). Besides atherothrombotic debris, unstable plaques
tion, the available anecdotal data comparing rotational atherec- contain a milieu of humoral factors that, if released down-
tomy use prior to drug-eluting stent (DES) deployment versus stream during percutaneous coronary interventions (PCI), have
DES alone does not support its systematic use (26). However, the potential for damaging the distal microcirculation. During
vigorous manipulation of DES through calcified lesions could STEMI, primary PCI of the thrombotic lesion requires particu-
potentially damage polymer coating integrity and lower DES lar care to reduce distal atherothrombotic embolization and its
efficiency. Although there are still some concerns regarding complications. The angiographic observance of distal emboli-
local delivery of drugs through a calcified lesion, clinical trials zation is rather common (*15%) during primary PCI and
with paclitaxel-eluting stent did not reveal differences in translates into poor left ventricular recovery and clinical out-
in-stent restenosis or thrombosis rates between calcified and come (40). Even in the absence of this complication and despite
noncalcified lesions (27). successful epicardial recanalization, inadequate myocardial
Cutting balloon angioplasty represents another alterna- reperfusion at the tissue level, reflected by incomplete ST-
tive to overcome fibrocalcified plaques (28,29). It consists of a segment resolution, is frequently observed immediately follow-
balloon catheter with three or four atherotomes or microsurgical ing the procedure (*40%) (41,42). Suboptimal myocardial
blades, which are three to five times sharper than conventional reperfusion at the end of the intervention is associated with
surgical blades (29). Atherotomes are placed longitudinally on reduced myocardial salvage and worse long-term outcome
the outer surface of a noncomplaint balloon. Radial expansion of than patients achieving adequate reperfusion (43,44). Lack of
the cutting balloon produces longitudinal incisions in the pla- myocardial reperfusion or no-reflow is due to mechanical
que and vessel, reducing tensile vessel stress (30) and elastic obstruction of the microcirculation, partly explained by the
recoil (31). Balloon sizes are 2.00 to 3.25 mm and its unique complex interplay of distal atherothromboic embolization, pos-
design protects the vessel from the edges of the atherotomes tischemic arteriolar damage, myocardial edema, reperfusion
when the balloon is deflated. The latter is vital to diminish the injury and the downstream release of vasoconstrictors and
[2/7/010/9:21:13] [482–491]
prothrombotic mediators (45). Although distal embolization is protocol, only patients with TIMI grade flow 0–2 were
a common phenomenon, the amount and composition of distal included. The study randomized 150 patients, and visible
debris varies largely. Consequently, a variety of techniques macroscopic atherothrombotic debris was noted by visual
have been designed to either reduce or prevent the amount of assessment in the filters in 48% of cases. Histopathologic
debris that migrates distally during the procedure. These analysis was performed in a subgroup of patients and particles
devices are divided into nonocclusive (distal embolic filter were recovered in all analyzed filters. The number of particles
protection) and occlusive (proximal or distal vessel occlusion) ranged from 8 to 48 per filter. Particles size ranged from 101 to
devices. 1299 mm in maximum diameter and from 212 to 1487 mm2 in
area. Most particles were composed of platelet clumps, red
cells, and fibrin, which led to the diagnosis of fresh thrombus
Distal Embolic Protection with Filter Devices (Fig. 37.3). All patients underwent 24-hour ST-segment mon-
These devices provide the ability to maintain distal perfusion
itoring to measure the extent of ST-segment recovery. In this
while protecting against macroembolization. Thus, they enable
trial, the adjunctive use of a distal filter device did not improve
antegrade flow and allow the operator to inject contrast during
myocardial reperfusion either by ST-segment resolution
the procedure. Nevertheless, several potential disadvantages
(Fig. 37.4) or by quantitative angiographic markers (TIMI
should be mentioned. Depending on the filter design, small
blush grade, TIMI flow, and TIMI frame count). Furthermore,
debris (<80–100 mm) and humoral mediators can go through
left ventricular function and six-month clinical outcome did not
the filter pores. Moreover, there is a potential risk of emboliza-
differ between groups. There was no specific subgroup dis-
tion while crossing the lesion with the device, especially if
cerned that derived benefit from adjunctive distal protection.
predilatation is required to advance the filter distally. Unlike
In a published meta-analysis (N ¼ 1467) including 8
proximal protection, distal filters do not protect side branches
randomized clinical trials, the role of distal protection with
proximal to the device. In the setting of STEMI, it can be
either filter or balloon occlusion during primary PCI did not
challenging to perform blinded distal filter placement in
show improvement in extent of reperfusion or 30-day mortality
patients with baseline epicardial TIMI flow 0–1.
when compared to PCI alone (Fig. 37.5) (52).
Several studies have shown the benefit of distal protec-
In summary, due to the lack of positive results and
tion in carotid stenting (46,47) and saphenous vein graft inter-
potential disadvantages, the routine use of distal filter protec-
vention (48,49). These positive results encouraged their use in
tion during primary PCI in native coronary arteries cannot be
the native coronary circulation among patients with acute cor-
recommended. Nonetheless, these studies have demonstrated
onary syndromes. Initial anecdotal data reported an improve-
that the use of distal filters appears to be feasible and safe.
ment in myocardial reperfusion parameters using different
Although it is not technically a filter device, the novel
surrogates such as myocardial blush grade or ST-segment
stent system MGuardTM (InspireMD, Tel Aviv, Israel, Fig. 37.6)
resolution (50).
(53) appears promising during intervention of very thrombotic
However, the PROMISE (Protection Devices in PCI-
or friable plaques and is worth mentioning. In addition to its
Treatment of Myocardial Infarction for Salvage of Endangered
low profile, MGuard has an ultrathin polymer mesh sleeve that
Myocardium) trial (51) utilized FilterWire EZTM Embolic Pro-
wraps around the stent, trapping plaque material behind the
tection System (Boston Scientific, California, U.S.) and random-
stent and reducing vessel trauma during stent deployment. In a
ized 200 patients undergoing PCI for STEMI and non-STEMI.
pilot primary PCI registry (n ¼ 60), angiographic distal embo-
Surrogate markers such as maximal adenosine-induced flow
lization was not detected, TIMI 3 flow and blush grade 3 were
velocity, infarct size, and incidence of distal embolization as
present in 100% and 77.8% of the patients, respectively (53).
well as 30-day mortality did not differ between the two groups.
Although positive results regarding myocardial reperfusion are
The inclusion of non-STEMI patients in this trial may have
encouraging, significant and meaningful data from a larger
blunted any differences between the two treatment groups.
randomized clinical trial are needed.
The PREMIAR (Protection of Distal Embolization in
High-Risk Patients with Acute ST-Segment Elevation Myocar-
dial Infarction) trial (41) was a prospective, randomized, con- Distal Embolic Protection with Occlusive Devices
trolled study that tested a low-profile distal protection filter This system is based on transient distal balloon occlusion
device (SpiderFX, ev3, Inc., Minnesota, U.S.; Fig. 37.2), during coupled with an aspiration catheter to prevent plaque debris
PCI in STEMI patients at high risk for embolic events. Thus, by embolization. Similar to distal filters, this system requires pas-
sage through the lesion and does not provide protection to
proximal side branches. In addition, it may not have an ade-
quately visible landing zone in the presence of baseline TIMI 0.
Unlike distal filters, balloon occlusion provokes myocardial
ischemia due to blood flow stagnation, and hence its duration
should remain as short as possible. Examples of distal occlud-
ing devices include PercuSurge GuardWire Distal Protection
Device (54) (Medtronic, Pennsylvania, U.S.) and TriActive
System (Kensey, Pennsylvania, U.S.).
The PercuSurge GuardWire Protection Device (Fig. 37.7)
occludes distally with a balloon while the Export catheter
(Medtronic) is used to aspirate the stagnated debris-containing
Figure 37.2 The SpiderFX Distal Filter (SpiderFX, ev3, Inc., blood. Thromboaspiration is performed after balloon dilation
Minnesota, U.S.). and stent deployment. This device was used with success in the
large, multicenter, randomized SAFER trial (Saphenous Vein
[2/7/010/9:21:13] [482–491]
Figure 37.3 (See color insert) Different views of particulates obtained from filters and panoramic views of a recent thrombus. An extensive
thrombus (A) is compared with very small particles (D). Some areas stained light pink (B, C) are suspicious of being plaque remnants.
Hematoxylin and eosin stain, 40. Source: Courtesy of Dr Jose Milei and From Ref. 41.
Figure 37.4 Complete ST-segment resolution (70%) after primary angioplasty in the PREMIAR and EMERALD trials. Source: From Refs.
41, 42.
Graft Angioplasty Free of Emboli Randomized). A total of 801 protection system was evaluated in the EMERALD (Enhanced
patients were randomized to saphenous vein graft intervention Myocardial Efficacy and Recovery by Aspiration of Liberalized
with and without the PercuSurge GuardWire Protection Device Debris) trial in the setting of STEMI. EMERALD randomized
System (55). The primary endpoint of the study, major adverse 501 patients who presented within six hours from symptom
cardiac event (MACE) rate at 30 days, was defined as the onset for primary (81%) or rescue PCI (19%) to receive PCI with
composite of death, MI, emergent bypass surgery, or target the GuardWire Plus (Medtronic) distal balloon occlusion sys-
vessel revascularization within 30 days of the index procedure. tem compared to angioplasty without distal protection (42).
There was a 6.9% absolute reduction (42% relative reduction) in Although visible debris were retrieved from 73% of patients
30-day MACE (9.6% for GuardWire vs. 16.5% for controls, p ¼ randomized to the device, this did not result in improved
0.001). This reduction was driven primarily by MI (8.6% vs. microvascular flow, reperfusion success, reduced infarct size,
14.7%, p ¼ 0.008) and no-reflow (3% vs. 9%, p ¼ 0.02). The same or enhanced event-free survival (Fig. 37.4).
[2/7/010/9:21:13] [482–491]
Figure 37.5 30-Day mortality outcomes according to the use of adjunctive distal protection, with odds ratios (OR) and 95% confidence
intervals (CI). Source: From Ref. 61.
Embolic Protection with Proximal Occlusion

Proximal occlusion devices cause blood flow stagnation by
occluding the target coronary artery “upstream” from the lesion.
With flow temporarily interrupted, thromboaspiration is per-
formed and the stent or balloon is delivered to the lesion.
Thromboaspiration is performed again and the proximal balloon
is deflated, which restores blood flow. Advantages of proximal
occlusion devices over distal occlusion or filters are several. First,
proximal occlusions do not need to cross the lesion, thereby
avoiding device-induced embolization. Second, device delivery
is not affected as much by tortuous vessels. Third, the landing
zone is always seen at the beginning of the procedure. Fourth,
side branches are protected and humoral mediators theoretically
do not reach the microcirculation. However, proximal occlusion
Figure 37.6 MGuardTM Coronary Stent System. The mesh sleeve devices may not have an adequate landing zone in ostial or
covering the stent allows less traumatic stent dilation and adequate proximal lesions.
entrapment of plaque debris behind the stent (InspireMD, Ltd., Tel The most studied proximal protection device is the Proxis
Aviv, Israel). Proximal Protection System Device (St. Jude Medical, Inc.,
Minnesota, U.S.; Fig. 37.8). A single-center STEMI registry (55)
Figure 37.7 (A) PercuSurge GuardWire Distal Protection Device.

PercuSurge, Minneapolis, Minnesota, U.S.. Note the distal balloon Figure 37.8 The Proxis Proximal Protection System (St. Jude
occluding the coronary vessel. (B) Export thromboaspiration cathe- Medical, Inc., Minnesota, U.S.) for proximal balloon occlusion and
ter (Medtronic, Minneapolis, Minnesota, U.S.). thromboaspiration.
[2/7/010/9:21:13] [482–491]
Figure 37.9 Two-year cumulative infarct-related artery

DES thrombosis rate for the overall population and
patients with large thrombus burden (LTB or grade 4)
and small thrombus burden (STB). Source: From Ref. 57.
(N ¼ 177) has shown encouraging results, with 96% final TIMI 3 28.2%) was associated with an improvement in myocardial
and 80% complete ST-segment resolution at 60 minutes, with a perfusion and a two-year survival benefit (92% vs. 83%, p ¼
30-day MACE rate of 4%. The PREPARE (Prospective Random- 0.051) (58) when compared to primary PCI alone. Thus, throm-
ized Trial of Proximal Microcirculatory Protection in Patients bus removal appears to confer short and long-term benefits,
with Acute Myocardial Infarction Undergoing Primary PCI) trial possibly avoiding the development of stent malapposition and
enrolled 284 primary PCI patients with the adjunctive use of the hence, reducing stent thrombosis rates. Again, however, pro-
Proxis device versus PCI alone. In this study, immediate perspective randomized clinical trials are needed to confirm this
centage ST-segment resolution was higher in the device group; hypothesis.
however, similar ST-segment resolution was seen at 30, 60, and Numerous thrombectomy devices have been tested dur-
90 minutes after the procedure. There was also no difference in ing primary PCI. Although there is ample heterogeneity in
30-day clinical outcome. According to the investigators (personal design, aspiration capacity, and operational principles, throm-
communication), there was a correlation between early ST-segment bectomy devices can be further divided into manual and
resolution and myocardial salvage and long-term outcomes. On the mechanical.
basis of these findings, a larger randomized trial is planned with
this device. Manual Thrombectomy Devices
Manual thrombectomy devices are user friendly; they have
minimal learning curve and do not impact overall procedural
Thrombectomy Devices time. The most studied and most widely used devices include
Thromboaspiration devices are used to remove thrombus from the Export Catheter (Medtronic, Minnesota, U.S.; Fig. 37.7B),
the coronary lesions during primary PCI to reduce distal Diver CE (Invatec, Roncadelle, Italy; Fig. 37.10), Rescue catheter
embolization. There is, however, another potential advantage (Boston Scientific), Pronto catheter (Vascular Solutions, Inc.,
of thrombus removal prior to stent implantation, possibly more Minneapolis, Minnesota, U.S.), and the QuickCat (Kensey).
so in the DES era. Following primary PCI, an intravascular Several small randomized studies with different manual
ultrasound study has detected high rates of late stent malap- devices have shown promising results. Recently, TAPAS (Throm-
position (56). This finding might partially explain the increased bus Aspiration during Percutaneous Coronary Intervention in
risk of stent thrombosis in this population when compared to
more elective procedures. Stent implantation may compress
thrombotic material behind the stent struts, which eventually
dissolves in the long term, rendering the stent malapposed. In a
primary PCI study (57), thrombus burden was graded from 0 to
4 according to thrombus length. The highest grade (grade 4)
was defined as the presence of thrombus length 2 times vessel
diameter. Patients with the highest thrombus grade had sig-
nificantly higher late stent thrombosis than patients with lower
thrombus grades (Fig. 37.9). In this retrospective study, the
adjunctive use of mechanical thrombectomy was independ-
ently associated with a reduction in the development of late
stent thrombosis. Further retrospective analysis performed by
these investigators only in patients with large thrombus burden Figure 37.10 Diver Catheter (Invatec, Roncadelle, Italy).
(n ¼ 266) showed that adjunctive thromboaspiration (72/266,
[2/7/010/9:21:13] [482–491]
Acute Myocardial Infarction Study) (59), a single-center

randomized trial (n ¼ 1071), evaluated the adjunctive use of
the Export catheter in an unselected STEMI population under-
going primary PCI. Manual thromboaspiration resulted in
better early myocardial perfusion, as assessed by the percent-
age of complete ST-segment resolution (Fig. 37.11), TIMI blush
grade, and TIMI 3 flow. This was associated with a significant
reduction in 12-month cardiac death (3.6% vs. 6.7%; OR 1.93;
95% CI 1.11–3.37; p ¼ 0.020) and reinfarction (2.2% vs. 4.3%;
p ¼ 0.05) (60). This trial is the first to report a significant
benefit in clinical hard endpoints with the use of an adjunctive
distal protection tool during primary PCI.
A meta-analysis including 9 trials and 2417 STEMI
patients undergoing primary PCI with or without adjunctive
manual thrombectomy was recently published (61). The use of
manual thromboaspiration significantly improved myocardial
reperfusion (Fig. 37.12B) and 30-day mortality (Fig. 37.12B).
Figure 37.11 ST-Segment Elevation Resolution in the TAPAS These results indicate that manual thromboaspiration offers a
trial. Source: From Ref. 59. significant clinical benefit prior to stenting during primary PCI,
and its use should be encouraged.
Figure 37.12 Final TIMI III flow (A) and 30-day mortality (B) rates according to the use of manual thrombectomy. Source: From Ref. 52.
[2/7/010/9:21:13] [482–491]
achieves a negative suction pressure of only 10 mmHg.

The more powerful suction due to the outflow pressure frag-
ments clots as they enter the catheter, hindering catheter
occlusion during clot aspiration. In experimental models,
clot mass removal is significantly higher using mechanical
compared to manual thromboaspiration (Fig. 37.14).
The multicenter AiMI (AngioJet Rheolytic Thrombec-
tomy in Patients Undergoing Primary Angioplasty for Acute
MI) trial (63) included 480 STEMI patients evaluating adjunc-
tive use of the Angiojet device during primary PCI versus
primary PCI alone. In this study, myocardial reperfusion and
myocardial salvage were worse in the thrombectomy group
when compared to PCI alone. These results further translated
into more 30-day adverse events in the thrombectomy group
(6.7% vs. 1.7%; p ¼ 0.01). To be fair, the control group
experienced a surprisingly low mortality rate (0.8% vs.
4.6%; p ¼ 0.02), much lower than normally observed in
STEMI trials. The data safety monitoring committee of the
Figure 37.13 The Angiojet Rheolytic Thrombectomy System AiMI trial felt that the deaths in the thrombectomy arm were
(Possis Medical, Inc., Minnesota, U.S.). not specifically caused by the device. However, ventricular
arrhythmias, hemolysis, and coronary perforation are poten-
tial complications associated with this device. This trial
included a high percentage of patients without angiographi-
cally visible thrombus. The use of this device may derive
Mechanical Thrombectomy Devices greater benefit in patients with large thrombus burden (58).
The Angiojet Rheolytic Thrombectomy System (62) (Possis The JETSTENT multicenter trial is enrolling STEMI patients
Medical, Inc., Minnesota, U.S.) creates an extremely high with high thrombus grade to PCI with the adjunctive Angiojet
negative pressure zone (600 mmHg) at its distal tip by thromboaspiration or PCI alone. Technical aspects related to
high-velocity saline jets that are directed back within the the use of Angiojet have also been modified to take the
catheter, resulting in removal of thrombus by suction through greatest advantage of this device and to diminish procedural
the outflow lumen, using the Venturi and Bernoulli effects risk. Until these study results are available, no clearly fore-
(Fig. 37.13). For comparison, manual thromboaspiration seeable recommendation for its use can be given.
Figure 37.14 (See color insert) Amount of clot mass removal with mechanical versus manual thrombectomy devices.
[2/7/010/9:21:13] [482–491]
CONCLUSION 20. MacIsaac AI, Bass TA, Buchbinder M, et al. High speed rotational
The majority of emboli protection devices require that each case atherectomy: outcome in calcified and noncalcified coronary
be carefully considered, balancing risk and benefit of the cho- artery lesions. J Am Coll Cardiol 1995; 26:731–736.
sen device. Such evaluation depends on patient’s demo- 21. Fischell TA, Haller S, Pulukurthy S, et al. Nicardipine and
adenosine "flush cocktail" to prevent no-reflow during rotational
graphics, clinical presentation, and coronary anatomy.
atherectomy. Cardiovasc Revasc Med 2008; 9:224–228.
Profound knowledge of the underlying coronary pathophysi- 22. Villanueva EV, Wasiak J, Petherick ES. Percutaneous transluminal
ology is crucial in determining the feasibility, safety, and ben- rotational atherectomy for coronary artery disease. Cochrane
efit of using any adjunctive interventional device. At this Database Syst Rev 2003:CD003334.
juncture, emboli protection devices are warranted in particular 23. Hong MK, Mintz GS, Lee CW, et al. Late stent malapposition after
cases of carotid and vein graft interventions. Manual aspiration drug-eluting stent implantation: an intravascular ultrasound anal-
thrombectomy catheters are effective in primary PCI cases. ysis with long-term follow-up. Circulation 2006; 113:414–419.
24. Siqueira DA, Abizaid AA, Costa Jde R, et al. Late incomplete appo-
sition after drug-eluting stent implantation: incidence and potential
REFERENCES for adverse clinical outcomes. Eur Heart J 2007; 28:1304–1309.
1. Libby P, Theroux P. Pathophysiology of coronary artery disease. 25. Cook S, Wenaweser P, Togni M, et al. Incomplete stent apposition
Circulation 2005; 111:3481–3488. and very late stent thrombosis after drug-eluting stent implanta-
2. Kruth HS. Sequestration of aggregated low-density lipoproteins tion. Circulation 2007; 115:2426–2434.
by macrophages. Curr Opin Lipidol 2002; 13:483–488. 26. Tran T, Brown M, Lasala J. An evidence-based approach to the use
3. Kruth HS, Huang W, Ishii I, et al. Macrophage foam cell forma- of rotational and directional coronary atherectomy in the era of
tion with native low density lipoprotein. J Biol Chem 2002; 277: drug-eluting stents: when does it make sense? Catheter Cardio-
34573–34580. vasc Interv 2008; 72:650–662.
4. Libby P. Inflammation in atherosclerosis. Nature 2002; 420:868–874. 27. Moussa I, Ellis SG, Jones M, et al. Impact of coronary culprit lesion
5. Swirski FK, Libby P, Aikawa E, et al. Ly-6Chi monocytes dominate calcium in patients undergoing paclitaxel-eluting stent implanta-
hypercholesterolemia-associated monocytosis and give rise to tion (a TAXUS-IV sub study). Am J Cardiol 2005; 96:1242–1247.
macrophages in atheromata. J Clin Invest 2007; 117:195–205. 28. Dahm JB, Topaz O, Woenckhaus C, et al. Laser-facilitated throm-
6. Yano T, Kawano H, Mochizuki H, et al. Atherosclerotic plaques bectomy: a new therapeutic option for treatment of thrombus-laden
composed of a large core of foam cells covered with thin fibrous coronary lesions. Catheter Cardiovasc Interv 2002; 56:365–372.
caps in twice-injured carotid arterial specimens obtained from high 29. Barath P, Fishbein MC, Vari S, et al. Cutting balloon: a novel
cholesterol diet-fed rabbits. J Atheroscler Thromb 2000; 7:83–90. approach to percutaneous angioplasty. Am J Cardiol 1991;
7. Cybulsky MI, Won D, Haidari M. Leukocyte recruitment to 68:1249–1252.
atherosclerotic lesions. Can J Cardiol 2004; 20(suppl B):24B–28B. 30. Liao CK, Bonneau HN, Yock PG, et al. Arterial response during
8. Yamawaki H, Pan S, Lee RT, et al. Fluid shear stress inhibits cutting balloon angioplasty: a volumetric intravascular ultrasound
vascular inflammation by decreasing thioredoxin-interacting pro- study. J Formos Med Assoc 2002; 101:756–761.
tein in endothelial cells. J Clin Invest 2005; 115:733–738. 31. Kawaguchi K, Kondo T, Shumiya T, et al. Reduction of early
9. Pasterkamp G, Galis ZS, de Kleijn DP. Expansive arterial remod- elastic recoil by cutting balloon angioplasty as compared to con-
eling: location, location, location. Arterioscler Thromb Vasc Biol ventional balloon angioplasty. J Invasive Cardiol 2002; 14:515–519.
2004; 24:650–657. 32. Bertrand OF, Bonan R, Bilodeau L, et al. Management of resistant
10. Krams R, Wentzel JJ, Oomen JA, et al. Shear stress in atherosclerosis, coronary lesions by the cutting balloon catheter: initial experience.
and vascular remodelling. Semin Interv Cardiol 1998; 3:39–44. Cathet Cardiovasc Diagn 1997; 41:179–184.
11. Slager CJ, Wentzel JJ, Gijsen FJ, et al. The role of shear stress in the 33. Dahm JB, Dorr M, Scholz E, et al. Cutting-balloon angioplasty
destabilization of vulnerable plaques and related therapeutic effectively facilitates the interventional procedure and leads to a
implications. Nat Clin Pract Cardiovasc Med 2005; 2:456–464. low rate of recurrent stenosis in ostial bifurcation coronary lesions:
12. Alfayoumi F, Srinivasan V, Geller M, et al. The no-reflow phe- A subgroup analysis of the NICECUT multicenter registry. Int J
nomenon: epidemiology, pathophysiology, and therapeutic Cardiol 2008; 124:345–350.
approach. Rev Cardiovasc Med 2005; 6:72–83. 34. Stone GW, Ellis SG, O’Shaughnessy CD, et al. Paclitaxel-eluting
13. Nakamura M, Nishikawa H, Mukai S, et al. Impact of coronary stents vs vascular brachytherapy for in-stent restenosis within
artery remodeling on clinical presentation of coronary artery bare-metal stents: the TAXUS V ISR randomized trial. JAMA 2006;
disease: an intravascular ultrasound study. J Am Coll Cardiol 295:1253–1263.
2001; 37:63–69. 35. Holmes DR Jr., Teirstein P, Satler L, et al. Sirolimus-eluting stents vs
14. Holmes DR Jr., Garratt KN, Topol EJ. Coronary angioplasty versus vascular brachytherapy for in-stent restenosis within bare-metal
excisional atherectomy trial: CAVEAT. Int J Cardiol 1992; 35:143–146. stents: the SISR randomized trial. JAMA 2006; 295:1264–1273.
15. Lefkovits J, Holmes DR, Califf RM, et al. Predictors and sequelae 36. Schoenhagen P, Stone GW, Nissen SE, et al. Coronary plaque
of distal embolization during saphenous vein graft intervention morphology and frequency of ulceration distant from culprit
from the CAVEAT-II trial. Coronary Angioplasty Versus Exci- lesions in patients with unstable and stable presentation. Arte-
sional Atherectomy Trial. Circulation 1995; 92:734–740. rioscler Thromb Vasc Biol 2003; 23:1895–1900.
16. Moussa I, Di Mario C, Moses J, et al. Coronary stenting after 37. Fuster V, Moreno PR, Fayad ZA, et al. Atherothrombosis and
rotational atherectomy in calcified and complex lesions. Angio- high-risk plaque: part I: evolving concepts. J Am Coll Cardiol
graphic and clinical follow-up results. Circulation 1997; 96:128–136. 2005; 46:937–954.
17. Meany TB, Friedman HZ, O’Neill WW. Coronary rotational 38. Virmani R, Burke AP, Farb A, et al. Pathology of the vulnerable
atherectomy: clinical application. J Invasive Cardiol 1991; 3:19–24. plaque. J Am Coll Cardiol 2006; 47:C13—C18.
18. Suguta M, Nakano A, Endo M, et al. Increase in serum troponin-I 39. Fukuda D, Shimada K, Tanaka A, et al. Comparison of levels of
following rotational atherectomy reliably predicts the occurrence of serum matrix metalloproteinase-9 in patients with acute myocar-
reversible wall motion abnormalities. Int J Cardiol 2006; 107:78–84. dial infarction versus unstable angina pectoris versus stable
19. Kini A, Reich D, Marmur JD, et al. Reduction in periprocedural angina pectoris. Am J Cardiol 2006; 97:175–180.
enzyme elevation by abciximab after rotational atherectomy of 40. Henriques JP, Zijlstra F, Ottervanger JP, et al. Incidence and clinical
type B2 lesions: Results of the Rota ReoPro randomized trial. Am significance of distal embolization during primary angioplasty for
Heart J 2001; 142:965–969. acute myocardial infarction. Eur Heart J 2002; 23:1112–1117.
[2/7/010/9:21:13] [482–491]
41. Cura FA, Escudero AG, Berrocal D, et al. Protection of Distal 53. Kaluski E, Hauptmann KE, Muller R, et al. Coronary stenting with
Embolization in High-Risk Patients with Acute ST-Segment Ele- MGuard: first-in-man trial. J Invasive Cardiol 2008; 20:511–515.
vation Myocardial Infarction (PREMIAR). Am J Cardiol 2007; 54. Stone GW, Rogers C, Hermiller J, et al. Randomized comparison
99:357–363. of distal protection with a filter-based catheter and a balloon
42. Stone GW, Webb J, Cox DA, et al. Distal microcirculatory pro- occlusion and aspiration system during percutaneous intervention
tection during percutaneous coronary intervention in acute of diseased saphenous vein aorto-coronary bypass grafts. Circu-
ST-segment elevation myocardial infarction: a randomized con- lation 2003; 108:548–553.
trolled trial. JAMA 2005; 293:1063–1072. 55. Koch KT, Haeck JD, Van Der Schaaf RJ, et al. Proximal embolic
43. De Luca G, Suryapranata H, de Boer MJ, et al. Combination of protection with aspiration in percutaneous coronary intervention
electrocardiographic and angiographic markers of reperfusion in using the Proxis device. Rev Cardiovasc Med 2007; 8:160–166.
the prediction of infarct size in patients with ST-segment elevation 56. Chechi T, Vittori G, Biondi Zoccai GG, et al. Single-center random-
myocardial infarction undergoing successful primary angioplasty. ized evaluation of paclitaxel-eluting versus conventional stent in
Int J Cardiol 2007; 117:232–237. acute myocardial infarction (SELECTION). J Interv Cardiol 2007;
44. Prasad A, Stone GW, Aymong E, et al. Impact of ST-segment 20:282–291.
resolution after primary angioplasty on outcomes after myocar- 57. Sianos G, Papafaklis MI, Daemen J, et al. Angiographic stent
dial infarction in elderly patients: an analysis from the CADIL- thrombosis after routine use of drug-eluting stents in ST-segment
LAC trial. Am Heart J 2004; 147:669–675. elevation myocardial infarction: the importance of thrombus bur-
45. Ito H, Maruyama A, Iwakura K, et al. Clinical implications of the den. J Am Coll Cardiol 2007; 50:573–583.
‘no reflow’ phenomenon. A predictor of complications and left 58. Sianos G, Papafaklis MI, Vaina S, et al. Rheolytic thrombectomy in
ventricular remodeling in reperfused anterior wall myocardial patients with ST-elevation myocardial infarction and large throm-
infarction. Circulation 1996; 93:223–228. bus burden: the Thoraxcenter experience. J Invasive Cardiol 2006;
46. Cremonesi A. The SPIDER Embolic Protection Device perfor- 18(suppl C):3C–7C.
mance evaluation in the carotid artery during percutaneous trans- 59. Svilaas T, van der Horst IC, Zijlstra F. Thrombus Aspiration
luminal angioplasty and or stenting. J Invasive Cardiol 2005; during Percutaneous coronary intervention in Acute myocardial
17:463–467. infarction Study (TAPAS)—study design. Am Heart J 2006; 151:
47. Zahn R, Ischinger T, Mark B, et al. Embolic protection devices for 597 e1–e7.
carotid artery stenting: is there a difference between filter and 60. Vlaar PJ, Svilaas T, van der Horst IC, et al. Cardiac death and
distal occlusive devices? J Am Coll Cardiol 2005; 45:1769–1774. reinfarction after 1 year in the Thrombus Aspiration during
48. von Korn H, Yu J, Huegl B, et al. Safety and efficacy of a new filter- Percutaneous coronary intervention in Acute myocardial infarc-
based protection system for aorto-coronary bypass graft intervention Study (TAPAS): a 1-year follow-up study. Lancet 2008;
tions: the ev3 Spider device. J Invasive Cardiol 2005; 17:352–355. 371:1915–1920.
49. Stone GW, Rogers C, Ramee S, et al. Distal filter protection during 61. De Luca G, Dudek D, Sardella G, et al. Adjunctive manual
saphenous vein graft stenting: technical and clinical correlates of thrombectomy improves myocardial perfusion and mortality in
efficacy. J Am Coll Cardiol 2002; 40:1882–1888. patients undergoing primary percutaneous coronary intervention
50. Limbruno U, Cortese B, Severi S. Thrombectomy for thrombus for ST-elevation myocardial infarction: a meta-analysis of random-
removal, filters for whatever else. J Cardiovasc Med (Hagerstown) ized trials. Eur Heart J 2008; 29:3002–3010.
2008; 9:408–409. 62. Antoniucci D. Rheolytic thrombectomy in acute myocardial
51. Guetta V, Mosseri M, Shechter M, et al. Safety and efficacy of the infarction: the Florence experience and objectives of the multi-
FilterWire EZ in acute ST-segment elevation myocardial infarc- center randomized JETSTENT trial. J Invasive Cardiol 2006;
tion. Am J Cardiol 2007; 99:911–915. 18(suppl C):32C–34C.
52. De Luca G, Suryapranata H, Stone GW, et al. Adjunctive mechan- 63. Ali A, Cox D, Dib N, et al. Rheolytic thrombectomy with percuta-
ical devices to prevent distal embolization in patients undergoing neous coronary intervention for infarct size reduction in acute
mechanical revascularization for acute myocardial infarction: a myocardial infarction: 30-day results from a multicenter random-
meta-analysis of randomized trials. Am Heart J 2007; 153:343–353. ized study. J Am Coll Cardiol 2006; 48:244–252.
[21/6/010/17:53:33] [492–503]
38
Complications of PCI: stent loss, coronary perforation and

aortic dissection
Amir-Ali Fassa and Marco Roffi
Various complications may occur during percutaneous coro- undeployed stent into the guiding catheter without having a
nary interventions (PCI) (Table 38.1). These may be related to proper alignment between the guiding catheter and the stent
the devices used during the procedure (e.g., coronary perfora- and the attempt to cross a calcified and tortuous lesion that has
tion, stent loss, or thrombosis) or to the intervention itself (e.g., not been sufficiently prepared with balloon predilatation or
abrupt vessel closure, distal embolization, side branch occlu- rotational atherectomy. Other mechanisms and favoring factors
sion). The circulatory system may also be affected in case of are listed in Table 38.2. Key points in preventing stent loss
aortic dissection or vascular or atheroembolic events. In addi- include adequate lesion preparation, sufficient guiding catheter
tion, access site complications may occur. Finally, PCI may be and guidewire support, verification of proper alignment with
complicated by systemic events in relation to adjunctive thera- the guiding catheter while retrieving an undelivered stent, use
pies and other aspects of the procedure (e.g., contrast nephr- of short and low-profile stents, and stenting of distal segments
opathy, hemorrhage, allergic reactions). These complications first if multiple stents are required.
may result in adverse events such as myocardial infarction,
renal failure, stroke, or death.
Despite the increasing procedural complexity and higher Outcome
risk profile of patients treated with PCI over the years, there has Stent loss is associated with a high rate of adverse events. In a
been a marked reduction of related in-hospital myocardial series of 387 procedures complicated by stent loss (from a total
infarction and mortality rates as well as of the need for emer- of 25,558 PCIs), Bolte et al. showed that stent embolization was
gency coronary artery bypass graft (CABG) surgery (1,2). This associated with an in-hospital mortality of 6% and a major
trend is likely the result of technological and pharmacological in-hospital adverse events rate of 25% (which included death,
advancements as well as of the increasing experience of oper- nonfatal myocardial infarction, CABG, and neurological deficit)
ators in management of high-risk situations and prevention (8). The complication rate was as high as 89% among patients
and treatment of procedural complications. with an undeployed stent loss in the coronary tree (N ¼ 47).
The present chapter focuses on three major complications of Conversely, more favorable outcomes were observed with
PCI, which include stent loss, coronary perforations, and aortic successful stent retrieval (N ¼ 111, complication rate of 9%),
dissection. Access site complications are described in chapter 8. peripheral stent embolization (N ¼ 157, complication rate of
15%), and successful intracoronary exclusion (crushing) or
deployment of the lost stent (N ¼ 72, complication rate of
STENT LOSS 31%). In a series of 20 patients by Eggbrecht et al., three patients
Introduction died following urgent CABG after stent embolization (mortality
Stents are currently used during 91% to 96% of PCIs (3,4). of 15%) (10). Similarly, Brilakis et al. reported increased rates of
Failure to deliver a stent to the target site occurs in 3.9% to 8.3% hospital mortality (3%) and emergency CABG (5%) following
of the procedures, mostly due to excessive tortuosity and/or stent embolization (11).
calcification of the target lesion or of the proximal vessel seg- These outcomes may or may not be directly related to
ments (5–7). As a consequence, the undeployed stent needs to stent embolization. For instance, maneuvers to retrieve a lost
be retrieved into the guiding catheter. This may lead to stent stent may result in coronary or peripheral artery injury. Like-
dislodgement from the delivery balloon catheter and subse- wise, an undeployed stent left in the coronary tree may also
quent stent embolization. According to earlier series, stent loss result in vessel thrombosis and myocardial infarction. More-
occurred in 0.9% to 3.4% of the procedures, while the incidence over, if stenting is applied to treat a dissection, failure to deliver
decreased to 0.3% to 0.5% in the most recent ones (6–16). This the stent at the target lesion may lead to abrupt vessel closure.
progress is likely due to the abandon of manually crimped Finally, stent loss likely identifies a population of patients with
stents; the improvement of stent profile, flexibility, and stent complex coronary anatomy and advanced disease at increased
adhesion to the balloon catheter; as well as the gained experi- risk of adverse clinical events, which may not be the direct
ence of operators with stent utilization (11). Importantly, the consequence of stent loss. As a general rule, peripheral stent
trend has been observed despite the increasing complexity of embolization is associated with a more favorable clinical course
coronary lesions treated by endovascular means (1,2). than coronary stent embolization, the latter rarely resulting in
Factors favoring stent loss include vessel anatomy, oper- limb ischemia or stroke (9–11). On the basis of the potentially
ator technique, and equipment characteristics. The most com- catastrophic complications of stent embolization, appropriate
mon situations associated with stent loss are the retraction of an management is of paramount importance.
[21/6/010/17:53:33] [492–503]
COMPLICATIONS OF PCI 493
Table 38.1 Complications of Percutaneous Coronary Intervention

Coronary complications
l Coronary perforation
l Stent loss
l Coronary dissection
l Abrupt vessel closure
l Distal embolization
l Side branch occlusion
l Stent thrombosis
Vascular complications
l Access site
Bleeding
Pseudoaneurysm
Arteriovenous fistula
Arterial thrombosis
Venous thrombosis
Infection Figure 38.1 Small-diameter balloon catheter technique for lost
l Retroperitoneal hematoma stent retrieval. A small-diameter balloon catheter is advanced on
l Dissection or perforation of peripheral vessels the guidewire toward the lost stent (upper diagram). The catheter is
l Iatrogenic aortic dissection advanced beyond the stent and inflated. Then, the balloon catheter
l Atheroembolism is pulled back, allowing retrieval of the lost stent (lower diagram).
Other
l Contrast nephropathy
l Gastrointestinal bleeding
l Radiation exposure
l Allergic reactions
l
Small-Diameter Balloon Catheter Technique
Ventricular arrhythmia, conduction disturbances
The use of small-diameter balloon catheter is probably the
simplest way to attempt stent retrieval. This technique may
be used if the stent gets stripped off the balloon but remains on
the guidewire. A small caliber balloon catheter (typically with a
Table 38.2 Conditions Associated with Coronary Stent Loss diameter of 1.0–1.5 mm) is passed through the undeployed
l Moderate to severe vessel calcification stent and is inflated distally. Subsequently, the balloon catheter,
l Proximal tortuosity/angulation the guidewire, the guiding catheter, and the stent are all
l Incomplete lesion preparation removed as a unit (Figs. 38.1 and 38.2) (11). To prevent possible
l Insufficient guiding catheter and/or guidewire support embolization to the cerebral circulation, the whole system
l Advancing or retrieving a stent through a previously placed one should always be pulled down to the descending aorta toward
l Retraction of undeployed stent into the guiding catheter the sheath before attempting to pull the stent-loaded balloon
l Use of longer stents catheter into the guiding catheter (9). Alternatively, some
l Use of manually crimped stents
authors have also described use of small-diameter balloon
catheters to attempt the advancement of the stent and the
deployment at the target lesion site (10).
Management Double-Wire Technique

Percutaneous treatments of lost coronary stents include The double-wire method is also a relatively straightforward
retrieval, exclusion from circulation (crushing), or deployment technique that can be used if the stent remains on the guide-
at the embolization site. To that purpose, several tools and wire. A second guidewire is advanced distal to the stent,
techniques have been described, such as the use of small- following which the two guidewires are twisted to intermingle
diameter balloon catheters, guidewires, vascular snares, vascu- their distal ends, therefore allowing entrapment of the stent,
lar baskets, vascular forceps/bioptomes, and dedicated frag- which can then be retrieved by pulling both wires (Fig. 38.3).
ment retrieval devices (Table 38.3). Retrieval success rates vary
widely, ranging from 29% to 100% (7–11,17). Loop Snare
If the stent is not on the guidewire anymore, a loop snare can be
used to retrieve the stent. Loop snares are either commercially
Table 38.3 Tools and Techniques for Percutaneous Retrieval of
Lost Coronary Stents
available (e.g., Microvena Amplatz Goose Neck Snare, Micro-
vena Corp., Plymouth, Minnesota, U.S.) or can be prepared de
l Low-profile/small balloon catheter technique novo in the catheterization laboratory using an exchange-length
l Two-wire technique 0.014-in. guidewire and a diagnostic multipurpose catheter.
l Loop snarea
l Biliary or myocardial biopsy forceps
Specific Vascular Retrieval Devices
l Retained fragment retrieverb
l Basket retrieval device The biliary or vascular forceps consists of a curved finger-like
projection that can be expanded/extended or contracted/
a
For example, Microvena Amplatz Goose Neck Snare (Microvena retracted, allowing entrapment of lost stent. The Cook retained
Corp., Plymouth, Minnesota, U.S.). fragment retriever (Cook Medical, Bloomington, Indiana, U.S.)
b
Cook Retrieval Set (Cook Medical, Bloomington, Indiana, U.S.). has an articulating arm operable from the proximal hub, which
[21/6/010/17:53:33] [492–503]
Figure 38.2 A stent was lost in the first obtuse marginal branch of the circumflex artery (arrowheads, left panel, left anterior oblique—caudal
view). A small balloon catheter was advanced through the stent (arrow, left panel), allowing successful retrieval of the lost stent in the absence
of complications, as documented in the final angiogram (right panel, left anterior oblique—caudal view).
deployment of the stent at the embolization site or exclusion

from the coronary circulation by crushing the undeployed stent
with an additional one. However, the two latter strategies
should be avoided if the stent is located in the left main
trunk or at other strategic locations (11). In fact, both an
undersized stent and an undeployed stent crushed against
the vessel wall put the patient at risk of subacute or late vessel
thrombosis. If none of these options are possible, efforts should
be made to at least move the stent into the aortic root, as
peripheral embolization usually has a more benign prognosis
than coronary embolization (8). Management options of periph-
eral stent embolization include retrieval with the small-diam-
eter balloon catheter or double-wire techniques or dedicated
tools mentioned above (snare, basket, forceps, Cook retrieval
set). While deployment of a coronary stent at a peripheral site is
Figure 38.3 Double-wire technique for lost stent retrieval. A sec-
rarely indicated due to the mismatch between the stent and the
ond guidewire is advanced distal to the lost stent (upper diagram).
The two guidewires are twisted to intermingle their distal ends,
vessel size, crushing the embolized device with a peripheral
allowing entrapment of the stent, which can then be retrieved by stent is a possible option. Because of the larger vessel size, this
pulling both wires (lower diagram). maneuver is far less likely to be associated with vessel throm-
bosis in the coronary tree. Overall, leaving an undeployed stent
in the peripheral circulation should be considered, as the tech-
nical difficulties and risk related to retrieval maneuvers may
allows grasping and retrieving of retained equipment frag- outweigh the risk of local complications related to the embol-
ments. The basket retrieval device (Cook Medical) consists of ized device (Fig. 38.5) (11). In addition, the location of the stent
several helically arranged loops that can be expanded or embolized in the peripheral circulation often remains
collapsed, allowing entrapment of the embolized stent, in par- unknown.
ticular if it has been deformed. These devices may not be If percutaneous efforts to treat the lost stent in the coro-
available in all cardiac catheterization laboratories, and their nary tree fail, surgical retrieval should usually be considered—
use should be reserved to experienced operators familiar with along with coronary bypass grafting if the target lesion is left
their use. Importantly, these devices cannot be used in the untreated—because of the intrinsic risk of vessel thrombosis.
coronary vasculature but are typically used to treat stent Conversely, surgical retrieval of a stent that has embolized in
embolization in large vessels such as the aorta, the iliac, and the peripheral circulation should be performed only in the
the femoral arteries (11). presence of symptoms resulting from device embolization.
The appropriate strategy for managing stent loss depends
on whether the stent has embolized in a coronary artery or in
the peripheral circulation and whether the stent is still on the Conclusions
guidewire or not (Fig. 38.4). If the stent has stripped off the Stent loss is a rare complication of PCI, which is associated with
delivery balloon catheter in the coronary artery but still remains a high rate of adverse outcomes if the stent is lost in the
on the guidewire, recovery with a small profile balloon catheter coronary tree. Stent embolization may be prevented by proper
or the double wire technique should be attempted. Use of a patient selection, equipment, and technique. Percutaneous treat-
loop snare should also be considered, especially if the guide ment of stent embolization, including retrieval, deployment at
wire is not on the stent anymore. Alternative options include embolization site, and exclusion are effective. Operators should
[21/6/010/17:53:33] [492–503]
Figure 38.4 Suggested algorithm for management of coronary stent loss.
CORONARY PERFORATIONS
Introduction
Coronary perforation is a rare but potentially life-threatening
complication that occurs in 0.2% to 0.9% of PCIs (18–32). With
atheroablative techniques such as directional or rotational athe-
rectomy and excimer laser angioplasty, the risk is somewhat
higher (0.3–3.3%) (19–21,33–35). The overall incidence of coro-
nary perforation during PCI has been unchanged over the last
20 years. This is likely explained by a balance between the
expected decrease due to changes in technique—such as the
more selective use of atheroablative techniques and the abandon
of high-pressure angioplasty with oversized compliant balloon
catheters—and the likely increase due to the greater number of
high-risk interventions and the use of stiff and hydrophilic
guidewires (20,21,36).
The diagnosis of coronary perforation is usually made
immediately by visualization of contrast extravasation on cor-
onary angiogram (36,37). However, in case of wire perforation,
the diagnosis may be made only later in the presence of
hemodynamic compromise and pericardial effusion on echo-
cardiography (18,23). Patients with frank perforation may com-
plain of severe chest pain, dizziness, nausea, and vomiting, out
Figure 38.5 X ray showing an embolized stent in the infrapopliteal of proportion to what typically observed during balloon infla-
circulation (arrow). No retrieval maneuver was attempted following tion. Heart rate usually increases and blood pressure falls, with
stent embolization. The patient remained asymptomatic during a rise in central venous pressure when tamponade develops. At
follow-up. times, vagal-mediated bradycardia may occur. ST segment
elevation or depression may ensue due to vessel occlusion at
the level of the perforation or distally (36).
be familiar with these techniques to properly manage this Ellis et al. proposed a classification of coronary perfora-
potentially serious complication. In case of failure to retrieve tion based on angiographic appearance (Table 38.4) (18). Type I
undeployed stents in the coronary vasculature, cardiac surgery perforations are characterized by a focal extraluminal crater
should be considered. Stent embolization in the peripheral limited to the media or adventitia in the absence of contrast
vasculature has a more benign prognosis and usually does not extravasation, and cannot be differentiated angiographically
require treatment. from the previously described NHLBI type C dissection,
[21/6/010/17:53:33] [492–503]
Table 38.4 Classification of Coronary Perforations and in particular of chronic total occlusion), lesion complexity
(type B2/C according to the ACC/AHA classification), and the
Type I Extraluminal crater without contrast extravasation
use of intravascular ultrasound (IVUS)-guided lumen optimi-
Type II Pericardial or myocardial blush without contrast
zation (20,21,24,41). The latter is probably explained by an
jet extravasation
Type III Extravasation through frank (1 mm) perforation incorrect interpretation of IVUS images to maximize the stent
Cavity spilling Perforation into an anatomic cavity (i.e., ventricle, lumen area (37). Moreover, advanced age, congestive heart
coronary sinus) failure, heavy calcification, extreme vessel tortuosity, small
vessel diameter, and balloon catheter or stent oversizing have
also been associated with an increased incidence of coronary
perforations. The use of stiff or hydrophilic guidewires has also
suggesting a continuum between dissection and perforation been reported in recent series as a cause of coronary perfora-
(38,39). Type II perforations include pericardial or myocardial tion, mostly in the distal bed (18–32). Cutting balloon angio-
blush without contrast jet extravasation (Fig. 38.6), while type plasty has also shown to be associated with a higher risk of
III perforations involve persistent extravasation with streaming coronary perforation that plain balloon angioplasty or stenting
or jet of contrast through a frank (1 mm) perforation (42). Following the widespread use of glycoprotein (GP) IIb/
(Fig. 38.7). If the jet is directed toward an anatomic cavity (e.g., IIIa inhibitors, concern of a potential increase in the occurrence
ventricle, coronary sinus), the perforation is classified as cavity of coronary perforation during administration of these potent
spilling (CS) type, type IV or alternatively type IIIB, as opposed drugs has been raised. However, several series have shown that
to type IIIA where the jet is directed toward the pericardium. GP IIb/IIIa inhibitors are not associated with a higher incidence
The most frequent type of perforation is type II (37–61%), while of coronary perforations. Nevertheless, coronary perforations
type I and III occur less often (18–44% and 19–36%, respec- occurring in patients under GP IIb/IIIa inhibitors are associated
tively) and type CS is rarely encountered (2–5%) (20,25,29). with a higher rate of adverse events, suggesting that these
Most type I and II perforations are caused by guidewires, while antiplatelet drugs may increase the harmful consequences but
type III are more frequently associated with use of stents or not necessarily the occurrence of coronary perforation. In this
atheroablative techniques (20,24,27). Muller et al. recently pro- respect, the association of GP IIb/IIIa inhibitors and hydro-
posed a modification of the Ellis classification, adding a type V philic wires appears particularly risky (20,23–25).
perforation describing distal coronary perforations by guide- The mainstay of perforation prevention is the cautious
wires (40). use of atheroablative techniques and of hydrophilic guidewires
Several retrospective series have sought to determine in complex anatomy, especially if there is concomitant admin-
predictive factors of coronary perforation among clinical, ana- istration of GP IIb/IIIa inhibitors. Of paramount importance is
tomical, and procedural characteristics (Table 38.5). The most a stable position of the guidewire, especially if it is hydrophilic,
consistent multivariate predictor of coronary perforation is the during advancement and retrieval of balloon catheters or
use of atheroablative techniques, which has been associated stents, as distal perforation is frequently the result of an
with an odds ratio for coronary perforation ranging from 2.7 to unnoticed wire movement. Finally, provided that the anatomy
6.8 in comparison to standard techniques (20,21,24). Additional is suitable, debulking with rotational atherectomy should be
independent predictors include female gender (likely due to favored for preparation of severely calcified lesions resistant to
the smaller size of the coronary arteries in women), history of conventional angioplasty over the use of oversized balloons
CABG (likely due to the higher prevalence of complex lesions inflated at high pressures.
Figure 38.6 Left panel: left anterior oblique view of a distal type II perforation of the left anterior descending coronary artery induced by a
hydrophilic guidewire, with pericardial extravasation (arrow). Right panel: right anterior oblique—cranial view of type II perforation, the left
circumflex coronary artery with myocardial staining following implantation of an oversized stent (arrowheads).
[21/6/010/17:53:33] [492–503]
that can often be managed conservatively or percutaneously,

the availability of covered stents, and the increasing experience
of operators (36).
Myocardial infarction, which can be due to vessel occlu-
sion, either as a direct consequence or as an intended treatment
of coronary perforation, may occur in 5% to 37% of cases
(25,28).
The mortality rate associated with coronary perforations
ranges from 0% to 17% (22,27). Predictors of mortality or major
adverse events include older age, female gender, chronic renal
failure, use of compliant balloons and treatment of angulated or
calcified lesions (19,21,27). Furthermore, type III perforations
are clearly associated with far worse outcomes than types I, II,
and CS (mortality of 19–22% for type III and <6% for types I, II,
and CS) (20,26).
Fasseas et al. reported that patients receiving GP IIb/IIIa
inhibitors at the time of coronary perforation more frequently
required the placement of a covered stent or emergency surgery
in comparison to patients not treated with these agents (20). A
Figure 38.7 Left lateral view of a type III perforation of the mid-left recent retrospective analysis has shown that guidewire-
anterior descending coronary artery following postdilatation of an induced coronary perforation was associated with a signifi-
implanted stent, with staining of the pericardial space. cantly lower incidence of major adverse cardiac events when
bivalirudin was used as an adjunctive therapy compared with
unfractionated heparin (30). Likewise, a pooled analysis from
three randomized controlled trials showed that outcomes fol-
Outcomes lowing coronary perforation with bivalirudin as an adjunctive
Coronary perforation is clearly associated with an increased therapy were not inferior in comparison to unfractionated
rate of in-hospital and long-term adverse events, which include heparin (31). These results suggest that bivalirudin may be as
myocardial infarction, emergency surgery, and death (21,31). safe as unfractionated heparin in the presence of a coronary
These outcomes are mostly the consequence of hemodynamic perforation. This is possibly explained by the short half-life of
compromise, resulting from pericardial effusion and tampo- the drug. However, in the presence of frank perforation, the
nade, or of ischemia due to failure to treat the target lesion or antithrombotic effect of bivalirudin cannot be immediately
vessel occlusion. According to the literature, tamponade may reversed while in patients treated with unfractionated heparin
complicate 12% to 55% of coronary perforations (20,22). How- protamine can be administered.
ever, the true incidence is likely lower because of underreport- Subendocardial or intramyocardial hematoma has been
ing of “benign” perforations. Tamponade usually appears reported as an extremely rare consequence of coronary perfo-
immediately after coronary perforation, although a late occur- ration. This dreadful complication is characterized by a bleed
rence has been reported in 21% to 52% of patients, with a delay that can progress in a relentless fashion, dissecting the epicar-
ranging from 2 to 10 hours following PCI (22,23,43–45). Tam- dium and the epicardial vessels from the underlying myocar-
ponade rarely develops in patients with prior CABG due to dium and avulsing perforator vessels, which in turn bleed
obliteration of the pericardial space by adhesions between the further, establishing a self-propagating process that may con-
epicardium and pericardium as well as the persistence of tinue even though the initial perforation site may have been
partial pericardiotomy following surgery (46). successfully sealed (47–49). It is believed that subendocardial
Emergency surgery, usually consisting of pericardial hematoma is more frequent in patients with prior CABG
drainage with or without CABG or repair or ligation of the because of adhesions between the epicardium and the pericar-
perforated artery, was reported in 24% to 41% of perforations in dium that can prevent blood from accumulating in the peri-
earlier series (18,34). More recent reports show a lower inci- cardial space. However, cases of subendocardial hematoma in
dence of emergency surgery (2–20%) (26,30). This reduction is patients without prior CABG have also been reported. Finally,
likely related to an improved early detection of coronary it is worth mentioning that bypass graft perforation may result
perforations, a decreased use of atheroablative devices, a higher in tamponade, mediastinal hemorrhage, hemoptysis, and sub-
proportion of guidewire-induced distal coronary perforations endocardial hematoma (25,46,48–50).
Table 38.5 Predictive Factors of Coronary Perforations

Clinical Anatomical Procedural
a
Female gender Complex lesions Atheroablative techniques
Chronic renal failure Chronic total occlusion Use of intravascular ultrasound
Prior CABG Heavy calcification Balloon catheter or stent oversizing
Elderly Extreme tortuosity Stiff or hydrophilic guidewires
Congestive heart failure Small vessel diameter Cutting balloon
Unstable angina/NSTEMI Graft anastomosis
a
ACC/AHA type B2/C (41).
Abbreviations: CABG, coronary artery bypass surgery; NSTEMI, non-ST segment elevation myocardial infarction.
[21/6/010/17:53:33] [492–503]
Management perforation, surveillance in a coronary care unit for at least

The first step in the treatment of coronary perforation is prompt 24 hours is indicated, as delayed tamponade may occur.
detection and classification of the type of perforation, as this
will subsequently dictate the management strategy (Fig. 38.8) Prolonged Balloon Inflation
(18). Type I perforations can generally be managed either In large type II and all type III perforations, the first maneuver,
conservatively or by the implantation of a conventional stent, even before starting cardiopulmonary resuscitation, pericardio-
based on the size and prognostic importance of the vessel centesis, or reversal of anticoagulation, should be the placement
(36,38). Similarly, type CS perforations may also be treated of a balloon catheter (with balloon to artery diameter ratio of
conservatively, with the vascular communication often closing 0.9–1.0) at the site of perforation or upstream in case of distal
spontaneously later on. Type II perforation management perforation, inflated at a low pressure (2–6 atm). Serial angio-
depends on the extent of the extravasation, as limited pericar- graphic assessments of the perforation should be performed at
dial and myocardial staining can usually be observed and 10 to 15 minutes inflation periods. In case of incomplete perfo-
treated conservatively, with discontinuation of anticoagulants ration sealing, the use of a perfusion balloon has been recom-
and/or GP IIb/IIIa inhibitors. Larger type II perforations mended to reduce myocardial ischemia, but the availability of
should be treated more aggressively, with prolonged balloon these devices in catheterization laboratories has markedly
inflation, reversal of anticoagulation and platelet transfusion if decreased (24,37,40).
indicated, and echocardiographic assessment of pericardial
effusion. Finally, type III perforations require immediate Cessation and Reversal of Adjunctive Therapy
actions with prolonged balloon inflation, discontinuation and If the patient is receiving GP IIb/IIIa inhibitors or bivalirudin,
reversal of antiplatelet and antithrombotic therapy (if possible), the drugs should be immediately discontinued in type II and III
hemodynamic supportive therapy, echocardiographic assess- perforations. Furthermore, in large type II and type III perfo-
ment of pericardial effusion and, if needed, pericardiocentesis, rations, anticoagulation by unfractionated heparin should be
and percutaneous or surgical treatment of perforation. In this reversed with administration of protamine sulfate to achieve an
respect, the balance between the ongoing ischemia (caused by activated clotting time of less than 150 seconds or a partial
the management of the perforation) and the hemodynamic thromboplastin time of less than 60 seconds. Treatment with
instability (caused by the perforation or by the ongoing ische- protamine is safe following stent implantation, without any
mia) is key (51). Following successful treatment of coronary increase in the incidence of stent thrombosis (52). If the patient
Figure 38.8 Suggested algorithm for the management of coronary perforations. Abbreviations: TTE, tranthoracic echocardiogram; CCU,
coronary care unit, CABG, coronary artery bypass grafting; CS, cavity spilling.
[21/6/010/17:53:33] [492–503]
was receiving abciximab, the platelet inhibitory function of this Emergency Surgery
potent drug can be reversed by platelet transfusions, as abcix- In the presence of persistent coronary perforation with hemo-
imab binds platelets with high affinity and has low free plasma dynamic compromise or large ischemic territory despite opti-
levels. Conversely, small molecules like tirofiban and eptifiba- mal nonoperative measures, surgical management should be
tide maintain high plasma concentration, and their antiplatelet considered. Emergency surgery can involve pericardial drain-
properties therefore remain unaffected by platelet transfusions. age, if percutaneous drainage failed or was insufficient, perfo-
However, in the presence of normal renal function, small ration repair or vessel ligation, and bypass grafting to vessels
molecules are cleared from plasma within few hours. In case with significant stenoses. Before transferring the patient to the
of renal failure, hemodialysis can be useful to diminish the operating room, a balloon catheter should be kept inflated at
effects of these drugs (24,37). Reversal of anticoagulation can- low pressure at the perforation site, as previously described,
not be achieved in patients under low molecular weight hep- and then removed during surgery.
arin, fondaparinux, or bivalirudin.
Conclusion
Hemodynamic Support Therapy Coronary perforation is a life-threatening complication of PCI,
Intravenous fluids should be administered immediately. Vaso- which usually occurs in high-risk anatomical settings or with
pressor and inotropic therapy as well as cardiopulmonary the use of atheroablative devices or hydrophilic wires. Prompt
resuscitation may become rapidly necessary following coronary recognition of this complication may be life saving, allowing for
perforation (24). Placement of an intra-aortic balloon pump effective percutaneous or surgical treatment of the perforation
may also be required in the presence of depressed left ven- as well as its hemodynamic consequences.
tricular function or large ongoing ischemia.
Pericardiocentesis AORTIC DISSECTION

The details of pericardiocentesis are described in chapter 14. If Introduction
a sizable coronary perforation is observed, an echocardiogram Iatrogenic aortic dissection (IAD) is an extremely rare compli-
should be performed emergently to assess the presence of cation of cardiac catheterization. Reported incidence ranges
pericardial effusion. In case of tamponade, pericardiocentesis from 2 to 4 cases per 10,000 cardiac catheterizations. The
is a life-saving maneuver. Once a multiple side-hole catheter is incidence is higher during PCI (3–15 cases per 10,000 proce-
placed in the pericardium and the hemorrhage has stopped, the dures) than during diagnostic coronary angiography (4–10
catheter should be maintained until the following day. In the cases per 100,000 catheterizations). This complication is more
absence of recurrent effusion on the echocardiogram, the cath- frequently observed in the setting of acute myocardial infarc-
eter can then be removed. Observation for an additional day in tion (19 cases per 10,000 procedures) (64–66). IAD is a retro-
a regular unit with a subsequent control ultrasound is war- grade extension into the aortic root of a dissection usually
ranted before discharge (24). located at the level of the coronary ostia, mainly caused by
guiding catheter manipulations (64). Following a coronary
artery dissection, contrast injections can cause retrograde pro-
Perforation Sealing gression of the flap up to the level of the coronary sinus of
Coronary perforations with persistent extravasation despite Valsalva, where the shearing forces of blood flow during
prolonged balloon inflation can successfully be treated with systole and diastole can further propagate the dissection in
implantation of polytetrafluoroethylene-covered stents (53–55). the aortic root (66,67). IAD without involvement of the coro-
The JOSTENT coronary stent graft (Abbott Vascular Devices, nary arteries has been reported in very rare instances due to
Abbott Park, Illinois, U.S.) consists of an ultrathin, biocompat- catheter manipulation in the aorta (68).
ible, and expandable polytetrafluoroethylene layer sandwiched The diagnosis of IAD is usually straightforward on
in between two coaxial balloon-expandable stents (for details angiography, characterized by contrast medium stagnating at
see chap. 32). Procedural success rates vary from 71% to 100%. the level of the aortic root or even extending to the ascending
The use of covered stents appears to have reduced the need for aorta. The spectrum of associated symptoms ranges from none
emergency surgery (55). However, due to their unfavorable to excruciating chest or back pain. Hypotension, hemodynamic
profile and stiffness, covered stents may not be delivered in compromise, and shock may ensue (69). On the basis of the
tortuous vessels or distal lesions. Importantly, the minimum extent of the aortic dissection, a classification of IAD was
vessel diameter that can accommodate a covered stent is 3.0 mm. proposed by Dunning et al. (64). Class I is defined as a focal
Covered stents should be placed either at the level of the dissection limited to the ipsilateral cusp of the dissected coro-
perforation or at a bifurcation upstream from perforation site nary artery. Class II and III include dissections that extend up
to seal the vessel. Dual antiplatelet therapy with aspirin and the aorta <40 mm and >40 mm, respectively (Table 38.6;
clopidogrel should also be continued during at least three Figs. 38.9 and 38.10).
months following covered stent implantation because of the
delayed reendothelialization with the risk of subacute stent
thrombosis. Occasional use of autologous venous-covered Table 38.6 Classification of Iatrogenic Coronary Dissections with
stents has also been described. The preparation is cumbersome Retrograde Extension into the Aortic Root
and unsuitable for emergency situations, although time inter- Class I Aortic dissection involving only the ipsilateral cusp
vals from vein harvest to stent deployment of 20 to 45 minutes Class II Aortic dissection involving cusp and extending up the
have been reported. Other reported percutaneous strategies for aorta <40 mm
treatment of distal coronary perforation include embolization Class III Aortic dissection involving cusp and extending up the
of coils, gelfoam, polyvinyl alcohol, thrombin, glue, or autolo- aorta >40 mm
gous blood clot (56–63). Source: From Ref. 64.
[21/6/010/17:53:33] [492–503]
aortic dissection such as Marfan syndrome or bicuspid aortic

valve would probably favor IAD, no apparent increase in
incidence of IAD has been be reported in such settings. The
most likely explanation is that the need to perform PCI on
Marfan patients is exceedingly rare. Likewise, although cases of
IAD in patients with cystic medial necrosis have been
described, it is believed that it does not represent a risk factor
for IAD, as low grades of degeneration are nonspecific and
associated with advanced age (64).
Outcomes
IAD can lead to extensive dissection involving the ascending
aorta up, the aortic arch, the supra-aortic vessels, and even the
descending aorta. Furthermore, extension of the intimal flap
toward the aortic valve may cause significant acute aortic
regurgitation as well as hemopericardium and tamponade.
Figure 38.9 Class II iatrogenic aortic dissection following retro- Emergency surgery is necessary in 6% to 33% of cases following
grade extension of a right coronary artery ostial dissection. IAD. Surgery will usually associate repair or replacement of the
ascending aorta with the treatment of the aortic valve, and
CABG or pericardial drainage if required. In-hospital mortality
rates after IAD may range between 0% and 25% (64–67,70).
IAD originates most frequently from the ostium of the
right coronary artery. In a review of 67 cases published in the Management
medical literature, Carstensen et al. reported that the dissection The management of IAD has not been yet standardized. A
spread from the right coronary ostium, left coronary ostium, commonly accepted management strategy is the immediate
and ostium of a saphenous vein graft in 87%, 12%, and 1% of stent implantation at the ostium of the coronary artery from
the cases, respectively (70). IAD occurs more often when deep where the intimal flap has spread followed by a conservative
intubation of the guiding catheter in the coronary artery is approach of watchful waiting for class I and II IAD, while class
required (e.g., in the treatment of chronic total occlusions). The III dissections should undergo immediate surgical treatment
use of left Amplatz guiding catheters also seems to be associ- (Fig. 38.11) (64). Some authors also recommend attempting
ated with a higher risk of IAD (64,65). As previously men- ostial stenting in patients with class III dissections (70). This
tioned, IAD appears to occur more frequently during PCI for strategy appears reasonable if the patient is in a relatively stable
acute myocardial infarction. This is most likely due to the hemodynamic condition and if stent implantation does not
urgency of the situation and operator haste in attempting to delay the surgical management. In all patients, unfractionated
rapidly achieve myocardial reperfusion, while an increased heparin should be reversed with protamine, and in patients
vulnerability of vessel walls in relation with the ongoing treated with abciximab, platelets should be administered. Peri-
inflammatory process cannot be excluded (64,66). Finally, cardiocentesis may be needed in patients with tamponade but
although underlying conditions that can cause spontaneous only if it is estimated that they may not survive the transfer to
Figure 38.10 Left lateral view of an aortic dissection following retrograde extension of a right coronary artery ostial dissection (arrow, left
panel), rapidly progressing to a class III dissection involving the entire aorta (arrowheads indicating dissection flap on aortography, right
panel). Note also the filling of the left ventricle as a sign of associated aortic regurgitation (arrows, right panel).
[21/6/010/17:53:33] [492–503]
Figure 38.11 Suggested algorithm for the management of iatrogenic aortic dissection. Abbreviations: CCU, coronary care unit; CABG,
coronary artery bypass grafting.
the operating room. For less unstable patients, surgical peri- 6. Nikolsky E, Gruberg L, Pechersky S, et al. Stent deployment
cardiocentesis in the operating theatre is preferred. Assessment failure: reasons, implications, and short- and long-term outcomes.
of the extension of the IAD can be achieved with an aortog- Catheter Cardiovasc Interv 2003; 59:324–328.
7. Cantor WJ, Lazzam C, Cohen EA, et al. Failed coronary stent
raphy. Additional diagnostic techniques for the acute phase
deployment. Am Heart J 1998; 136:1088–1095.
include transthoracic and transoesophageal echocardiography
8. Bolte J, Neumann U, Pfafferott C, et al. on behalf of the Arbeits-
and computed tomography. gemeinschaft Leitende Kardiologische Krankenhausärtze (ALKK).
Incidence, management, and outcome of stent loss during intra-
coronary stenting. Am J Cardiol 2001; 88:565–567.
Conclusions 9. Alfonso F, Martinez D, Hernández R, et al. Stent embolization
IAD is a rare complication of cardiac catheterization, which during intracoronary stenting. Am J Cardiol 1996; 78:833–835.
occurs more frequently in the setting of PCI for acute myocar- 10. Eggebrecht H, Haude M, von Birgelen C, et al. Nonsurgical
dial infarction. Patients with a dissection confined to the aortic retrieval of embolized coronary stents. Catheter Cardiovasc Interv
sinus of Valsalva or extending less than 40 mm up the ascend- 2000; 51:432–440.
ing aorta should be treated with stenting of the coronary ostium 11. Brilakis ES, Best PJ, Elesber AA, et al. Incidence, retrieval methods,
and then monitored in the intensive care unit, while dissections and outcomes of stent loss during percutaneous coronary inter-
extending more than 40 mm up the aorta should be treated vention: a large single-center experience. Catheter Cardiovasc
Interv 2005; 66:333–340.
with immediate surgery.
12. Dunning DW, Kahn JK, O’Neill WW. The long-term consequences
of lost intracoronary stents. J Interv Cardiol 2002; 15:345–348.
13. Kozman H, Wiseman AH, Cook JR. Long-term outcome following
REFERENCES coronary stent embolization or misdeployment. Am J Cardiol
1. Venkitachalam L, Kip KE, Selzer F, et al. Twenty-year evolution of 2001; 88:630–634.
percutaneous coronary intervention and its impact on clinical 14. Kammler J, Leisch F, Kerschner K, et al. Long-term follow-up in
outcomes: a report from the National Heart, Lung, and Blood patients with lost coronary stents during interventional proce-
Institute-Sponsored, Multicenter 1985–1986 PTCA and 1997–2006 dures. Am J Cardiol 2006; 98:367–369.
Dynamic Registries. Circ Cardiovasc Interv 2009; 2:6–13. 15. Colkesen AY, Baltali M, Acil T, et al. Coronary and systemic stent
2. Singh M, Rihal CS, Gersh BJ, et al. Twenty-five-year trends in embolization during percutaneous coronary interventions: a sin-
in-hospital and long-term outcome after percutaneous coronary gle center experience. Int Heart J 2007; 48:129–136.
intervention: a single-institution experience. Circulation 2007; 16. Roffi M, Luscher TF, Sutsch G, et al. Failure to retrieve unde-
115:2835–2841. ployed paclitaxel-eluting coronary stents. Am J Cardiol 2006;
3. Wang TY, Peterson ED, Dai D, et al.; National Cardiovascular 97:502–505.
Data Registry. Patterns of cardiac marker surveillance after elec- 17. Foster-Smith KW, Garratt KN, Higano ST, et al. Retrieval techni-
tive percutaneous coronary intervention and implications for the ques for managing flexible intracoronary stent misplacement.
use of periprocedural myocardial infarction as a quality metric: a Cathet Cardiovasc Diagn 1993; 30:63–68.
report from the National Cardiovascular Data Registry (NCDR). 18. Ellis SG, Ajluni S, Arnold AZ, et al. Increased coronary perforation
J Am Coll Cardiol 2008; 51:2068–2074. in the new device era. Incidence, classification, management, and
4. Williams DO, Abbott JD, Kip KE; DEScover Investigators. Out- outcome. Circulation 1994; 90:2725–2730.
comes of 6906 patients undergoing percutaneous coronary inter- 19. Gruberg L, Pinnow E, Flood R, et al. Incidence, management, and
vention in the era of drug-eluting stents: report of the DEScover outcome of coronary artery perforation during percutaneous cor-
Registry. Circulation 2006; 114:2154–2162. onary intervention. Am J Cardiol 2000; 86:680–682.
5. Lohavanichbutr K, Webb JG, Carere RG, et al. Mechanisms, man- 20. Fasseas P, Orford JL, Panetta CJ, et al. Incidence, correlates,
agement, and outcome of failure of delivery of coronary stents. management, and clinical outcome of coronary perforation: anal-
Am J Cardiol 1999; 83:779–781. ysis of 16,298 procedures. Am Heart J 2004; 147:140–145.
[21/6/010/17:53:33] [492–503]
21. Stankovic G, Orlic D, Corvaja N, et al. Incidence, predictors, 41. Smith SC Jr., Dove JT, Jacobs AK, et al. ACC/AHA Guidelines for
in-hospital, and late outcomes of coronary artery perforations. Percutaneous Coronary Intervention (Revision of the 1993 PTCA
Am J Cardiol 2004; 93:213–216. Guidelines)-Executive Summary. Circulation 2001; 103:3019–3041.
22. Fukutomi T, Suzuki T, Popma JJ, et al. Early and late clinical 42. Mauri L, Bonan R, Weiner BH, et al. Cutting balloon angioplasty
outcomes following coronary perforation in patients undergoing for the prevention of restenosis: results of the Cutting Balloon
percutaneous coronary intervention. Circ J 2002; 66:349–356. Global Randomized Trial. Am J Cardiol 2002; 90:1079–1083.
23. Gunning MG, Williams IL, Jewitt DE, et al. Coronary artery 43. Von Sohsten R, Kopistansky C, Cohen M, et al. Cardiac tamponade
perforation during percutaneous intervention: incidence and out- in the “new device” era: evaluation of 6999 consecutive percuta-
come. Heart 2002; 88:495–498. neous coronary interventions. Am Heart J 2000; 140:279–283.
24. Dippel EJ, Kereiakes DJ, Tramuta DA, et al. Coronary perforation 44. Fejka M, Dixon SR, Safian RD, et al. Diagnosis, management, and
during percutaneous coronary intervention in the era of abciximab clinical outcome of cardiac tamponade complicating percutaneous
platelet glycoprotein IIb/IIIa blockade: an algorithm for percuta- coronary intervention. Am J Cardiol 2002; 90:1183–1186.
neous management. Catheter Cardiovasc Interv 2001; 52:279–286. 45. Ajluni SC, Glazier S, Blankenship L, et al. Perforations after percu-
25. Witzke CF, Martin-Herrero F, Clarke SC, et al. The changing taneous coronary interventions: clinical, angiographic, and thera-
pattern of coronary perforation during percutaneous coronary peutic observations. Cathet Cardiovasc Diagn 1994; 32:206–212.
intervention in the new device era. J Invasive Cardiol 2004; 46. Lowe R, Hammond C, Perry RA. Prior CABG does not prevent
16:257–301. pericardial tamponade following saphenous vein graft perforation
26. Ramana RK, Arab D, Joyal D, et al. Coronary artery perforation associated with angioplasty. Heart 2005; 91:1052.
during percutaneous coronary intervention: incidence and out- 47. Rahman N, Sharif H, Jafary FH. Intramyocardial hematoma after
comes in the new interventional era. J Invasive Cardiol 2005; coronary perforation during percutaneous coronary intervention-
17:603–605. anticipated and treated. J Invasive Cardiol 2008; 20:E224–E228.
27. Javaid A, Buch AN, Satler LF, et al. Management and outcomes of 48. Quan VH, Stone JR, Couper GS, et al. Coronary artery perforation
coronary artery perforation during percutaneous coronary inter- by cutting balloon resulting in dissecting subepicardial hematoma
vention. Am J Cardiol 2006; 98:911–914. and avulsion of the vasculature. Catheter Cardiovasc Interv 2005;
28. Eggebrecht H, Ritzel A, von Birgelen C, et al. Acute and long-term 6:163–168.
outcome after coronary artery perforation during percutaneous 49. Shekar PS, Stone JR, Couper GS. Dissecting sub-epicardial hema-
coronary interventions. Z Kardiol 2004; 93:791–798. toma-challenges to surgical management. Eur J Cardiothorac Surg
29. Kiernan TJ, Yan BP, Ruggiero N, et al. Coronary artery perfora- 2004; 26:850–853.
tions in the contemporary interventional era. J Interv Cardiol 2009; 50. Shammas NW, Thondapu VR, Winniford MD, et al. Perforation of
22:350–353. saphenous vein graft during coronary stenting: a case report.
30. Kini AS, Rafael OC, Sarkar K, et al. Changing outcomes and Cathet Cardiovasc Diagn 1996; 38:274–276.
treatment strategies for wire induced coronary perforations in 51. Satler LF. A revised algorithm for coronary perforation. Catheter
the era of bivalirudin use. Catheter Cardiovasc Interv 2009; Cardiovasc Interv 2002; 57:215–216.
74:700–707. 52. Briguori C, Di Mario C, De Gregorio J, et al. Administration of
31. Doll JA, Nikolsky E, Stone GW, et al. Outcomes of patients with protamine after coronary stent deployment. Am Heart J 1999;
coronary artery perforation complicating percutaneous coronary 138:64–68.
intervention and correlations with the type of adjunctive antith- 53. Lansky AJ, Yang YM, Khan Y, et al. Treatment of coronary artery
rombotic therapy: pooled analysis from REPLACE-2, ACUITY, perforations complicating percutaneous coronary intervention
and HORIZONS-AMI trials. J Interv Cardiol 2009; 22:453–459. with a polytetrafluoroethylene-covered stent graft. Am J Cardiol
32. Shimony A, Zahger D, Van Straten M, et al. Incidence, risk factors, 2006; 98:370–374.
management and outcomes of coronary artery perforation during 54. Ly H, Awaida JP, Lespérance J, et al. Angiographic and clinical
percutaneous coronary intervention. Am J Cardiol 2009; 104: outcomes of polytetrafluoroethylene-covered stent use in signifi-
1674–1677. cant coronary perforations. Am J Cardiol 2005; 95:244–246.
33. Bittl JA, Ryan TJ, Keaney JF, et al. Coronary artery perforation 55. Briguori C, Nishida T, Anzuini A, et al. Emergency polytetra-
during excimer laser coronary angioplasty. The percutaneous fluoroethylene-covered stent implantation to treat coronary rup-
Excimer Laser Coronary Angioplasty Registry. J Am Coll Cardiol tures. Circulation 2000; 102:3028–3031.
1993; 21:1158–1165. 56. Colombo A, Itoh A, Di Mario C, et al. Successful closure of a
34. Holmes DR Jr., Reeder GS, Ghazzal ZM, et al. Coronary perfora- coronary vessel rupture with a vein graft stent: case report. Cathet
tion after excimer laser coronary angioplasty: the Excimer Laser Cardiovasc Diagn 1996; 38:172–174.
Coronary Angioplasty Registry experience. J Am Coll Cardiol 57. Caputo RP, Amin N, Marvasti M, et al. Successful treatment of a
1994; 23:330–335. saphenous vein graft perforation with an autologous vein-covered
35. Cohen BM, Weber VJ, Relsman M, et al. Coronary perforation stent. Catheter Cardiovasc Interv 1999; 48:382–386.
complicating rotational ablation: the U.S. multicenter experience. 58. Cordero H, Gupta N, Underwood PL, et al. Intracoronary autol-
Cathet Cardiovasc Diagn 1996; (suppl 3):55–59. ogous blood to seal a coronary perforation. Herz 2001; 26:157–160.
36. Klein LW. Coronary artery perforation during interventional 59. Störger H, Ruef J. Closure of guide wire-induced coronary artery
procedures. Catheter Cardiovasc Interv 2006; 68:713–717. perforation with a two-component fibrin glue. Catheter Cardio-
37. Nair P, Roguin A. Coronary perforations. EuroIntervention 2006; vasc Interv 2007; 70:237–240.
2:363–370. 60. Assali AR, Moustapha A, Sdringola S, et al. Successful treatment
38. Rogers JH, Lasala JM. Coronary artery dissection and perforation of coronary artery perforation in an abciximab-treated patient
complicating percutaneous coronary intervention. J Invasive Car- by microcoil embolization. Catheter Cardiovasc Interv 2000; 51:
diol 2004; 16:493–499. 487–489.
39. Coronary artery angiographic changes after PTCA. In: Manual of 61. Yoo BS, Yoon J, Lee SH, et al. Guidewire-induced coronary artery
Operations: NHLBI PTCA Registry. Bethesda, MD: National perforation treated with transcatheter injection of polyvinyl alco-
Heart, Lung, and Blood Institute, 1985:6–9. hol form. Catheter Cardiovasc Interv 2001; 52:231–234.
40. Muller O, Windecker S, Cuisset T, et al. Management of two major 62. Fischell TA, Korban EH, Lauer MA. Successful treatment of distal
complications in the cardiac catheterisation laboratory: the no- coronary guidewire-induced perforation with balloon catheter
reflow phenomenon and coronary perforations. EuroIntervention delivery of intracoronary thrombin. Catheter Cardiovasc Interv
2008; 4:181–183. 2003; 58:370–374.
[21/6/010/17:53:33] [492–503]
63. Dixon SR, Webster MW, Ormiston JA, et al. Gelfoam embolization 67. Pérez-Castellano N, Garcı́a-Fernández MA, Garcı́a EJ, et al. Dis-
of a distal coronary artery guidewire perforation. Cardiovasc section of the aortic sinus of Valsalva complicating coronary
Interv 2000; 49:214–217. catheterization: cause, mechanism, evolution, and management.
64. Dunning DW, Kahn JK, Hawkins ET, et al. Iatrogenic coronary Cathet Cardiovasc Diagn 1998; 43:273–279.
artery dissections extending into and involving the aortic root. 68. Wyss CA, Steffel J, Lüscher TF. Isolated acute iatrogenic aortic
Catheter Cardiovasc Interv 2000; 51:387–393. dissection during percutaneous coronary intervention without
65. Gómez-Moreno S, Sabaté M, Jiménez-Quevedo P, et al. Iatrogenic involvement of the coronary arteries. J Invasive Cardiol 2008;
dissection of the ascending aorta following heart catheterisation: 20:380–382.
incidence, management and outcome. EuroIntervention 2006; 69. Januzzi JL, Sabatine MS, Eagle KA, et al. International Registry of
2:197–202. Aortic Dissection Investigators. Iatrogenic aortic dissection. Am J
66. Yip HK, Wu CJ, Yeh KH, et al. Unusual complication of retrograde Cardiol 2002; 89:623–626.
dissection to the coronary sinus of Valsalva during percutaneous 70. Carstensen S, Ward MR. Iatrogenic aortocoronary dissection: the
revascularization: a single-center experience and literature review. case for immediate aortoostial stenting. Heart Lung Circ 2008;
Chest 2001; 119:493–501. 17:325–329.
[17/6/010/8:25:34] [504–514]
39
Intra-aortic balloon pump counterpulsation,

percutaneous left ventricular support
Troy Weirick, Wendell Ellis, and Richard W. Smalling
INTRODUCTION AMI were 10 times more likely to die following PTCA when
Currently, no formal guidelines detail the indications for tem- compared to patients presenting with uncomplicated disease.
porary left ventricular (LV) support. This lack of consensus In a worst-case scenario, the combination of low EF and AMI,
notwithstanding, short-term mechanical cardiac assistance may periprocedural mortality was 50 times greater in high-risk
be appropriate in a few well-defined circumstances. In the patients (9). Around the same time, Ellis et al. showed that
setting of acute myocardial infarction (AMI) with severely mortality related to acute vessel closure during PCTA was
compromised ventricular function, temporary mechanical sup- largely a function of the area of myocardium at risk (10).
port unloads the left ventricle and augments cardiac output. These studies and other early work with supported PTCA
Unloading the ventricle, particularly when initiated prior to demonstrated a role for temporary LV assistance during high-
reperfusion, may improve myocardial blood flow and reduces risk procedures.
reperfusion injury. Improving cardiac output mitigates end-
organ damage potentially limiting the systemic inflammatory
response. In the case of chronic heart failure, temporary LV ANATOMIC CONSIDERATIONS
support reverses the acute metabolic derangements associated The primary anatomic considerations limiting percutaneous LV
with decompensation. If the ventricle is able to recover suffi- support are access site disease, aortic valve dysfunction, and
ciently, medical management may be resumed. If ventricular integrity of the thoracic and abdominal aorta. All of the devices
function remains inadequate, temporary support may serve as detailed in this chapter require relatively large arterial access
a bridge to more definitive therapy. Finally, during high-risk conduits. With the exception of sheathless intra-aortic balloon
percutaneous coronary intervention (PCI), temporary support pump (IABP) insertion, all forms of percutaneous LV support
provides circulatory protection in the event of acute vessel described require a minimum of 7.5-Fr arterial access. Further-
closure or catastrophic LV compromise. more, there is a significant trade-off between the degree of LV
Cardiogenic shock occurs in approximately 7% of support provided and the size of arterial cannula required. The
patients hospitalized for acute STEMI and is the leading IABP offers minimal augmentation of cardiac output (0.5 L/
cause of early death in post-MI patients (1). Despite maximal min), but sheathed insertion is nominal at 7.5 Fr. The Tandem-
medical management, including early thrombolysis and intra- Heart offers the greatest cardiac assistance, 4.0 L/min, but
aortic balloon counterpulsation (IABC), 30-day mortality from requires 21-Fr venous access and unilateral 15- to 17-Fr or
cardiogenic shock following AMI remains unacceptably high at bilateral 12-to 15-Fr arterial access. The Impella LP2.5 is inter-
roughly 50% (2). A strategy of early, aggressive PCI appears to mediate, requiring a single 13- to 14-Fr arterial cannula and
have a positive, sustained influence at one-year and on late offering up to 2.5 L/min of flow. Because of the prevalence of
survival (3,4). However, even with successful mechanical significant peripheral vascular disease in our patients, if vessel
reperfusion, viable myocardium may be stunned and noncon- anatomy is unknown at the time of intervention, angiography
tractile for up to a week (5). Without adequate LV support of the aorta with runoff is generally performed prior to device
during this period of myocardial recovery, the cycle of shock insertion.
and further ischemic damage goes unchecked, leading to more Device selection is also limited by aortic valve disorders
profound pump failure, propagation of the systemic inflamma- and diseases of the aorta. Moderate to severe aortic regurgita-
tory response syndrome, and eventually death (6). tion is a contraindication to Impella, TandemHeart, and IABP
The safety and durability of PCI has improved dramat- support. All of these devices can potentially worsen the sever-
ically in the past 10 years due to better antiplatelet agents and ity of aortic regurgitation negating the benefits of LV assistance.
improved stent technology. Attendant reductions in periproce- In addition, due to its transvalvular design, the Impella cannot
dural complications, a decreased need for target lesion revas- be used in patients with moderate to severe aortic stenosis or in
cularization and mortality comparable to conventional bypass individuals with mechanical aortic valves. When significant
grafting has lead to greater acceptance of more aggressive PCI valvular disease is suspected, echocardiography prior to arrival
strategies (7,8). Despite noteworthy procedural success, early in the catheterization laboratory is highly recommended.
attempts at percutaneous transluminal coronary angioplasty Lastly, severe atherosclerotic disease or significant aneurysm
(PTCA) in high-risk individuals demonstrated unacceptable in- of the thoracoabdominal aorta is a contraindication to both
hospital mortality and required frequent target lesion revascu- IABC and Impella support. Repeated balloon inflations and the
larization. According to work performed by Hartzler and associated mechanical stress may lead to dislodgement and
colleagues during the prestent era, individuals presenting embolization of atherosclerotic debris. The same repetitive
with left main disease, a low ejection fraction (EF) or with forces combined with systemic anticoagulation may lead to
[17/6/010/8:25:34] [504–514]
IABP, LV SUPPORT 505
devastating rupture of aortic aneurysms. Patients with severe link and colleagues directly measured coronary artery blood
disease of the aorta have generally been excluded from previ- flow in 11 Impella-assisted patients undergoing high-risk PCI.
ous Impella trials; therefore, the safety of the Impella in this In this study, investigators found that as Impella support
population is uncertain. increased, hyperemic coronary flow velocity increased as
well. These investigators also noted an inverse relationship
between Impella support and coronary microvascular resis-
FUNDAMENTALS tance (17). This combination of improved systemic hemody-
Two basic concepts provide the rationale for LV assistance namics along with superior myocardial oxygen supply and
during AMI, cardiogenic shock, and high-risk PCI. First, aug- decreased metabolic demand compose a compelling argument
mentation of cardiac output enhances systemic hemodynamics for mechanical LV support during AMI.
and improves end-organ perfusion. Second, LV unloading
modestly improves coronary artery blood flow while reducing
myocardial oxygen demand, thus limiting adverse remodeling. INDICATIONS
Animal experiments and human trials have demonstrated Clinical applications of the IABP and TandemHeart are fairly
superior hemodynamics and improved LV unloading with well established and generally similar. As a much newer
mechanical LV support. An early trial with the Hemopump device, indications for the Impella LP2.5 are evolving. On the
(Medtronic Inc., Minneapolis, Minnesota, U.S.) elegantly dem- basis of data from the Benchmark Registry, the most common
onstrated enhanced systemic hemodynamics in patients suffer- applications of IABC are hemodynamic support during cardiac
ing from cardiogenic shock (11). In this trial, 53 patients catheterization procedures (21%), cardiogenic shock (19%),
meeting criteria for cardiogenic shock (cardiac index <2.0 L/ weaning from bypass (16%), perioperative support in high-
min/m2, pulmonary capillary wedge pressure >18 mmHg, risk patients (13%), and refractory unstable angina (12%) (18).
systolic blood pressure <90 mmHg) were assigned to Hemo- Common indications for TandemHeart support include AMI,
pump insertion. In the 41 patients with successful placement, postcardiotomy pump failure, decompensated heart failure,
cardiac index increased from 1.6 to 2.2 L/min/m2, pulmonary transplant allograft dysfunction, cardiac arrest, and right ven-
capillary wedge pressure decreased from 27 to 17 mmHg, and tricular infarction. The Impella has demonstrated effectiveness
mean arterial pressure increased from 56 to 67 mmHg. Later, in in support during high-risk PCI, cardiogenic shock, and MI.
a sheep model of anterior MI, Meyns et al. demonstrated that Table 39.1 provides a more complete list of indications for
the Impella 5.0 improved cardiac output and increased mean temporary cardiac support.
arterial pressure while decreasing myocardial oxygen con-
sumption (12). In this trial, fully supported animals demon-
strated lower LV end diastolic pressure during ischemia and EQUIPMENT
reperfusion compared to the unsupported group. Furthermore, Currently there is no ideal device for the acutely decompen-
despite increased mean arterial pressure, device-supported sated ventricle; however, a great deal has been learned from
animals demonstrated a lower dP/dt max and reduced myo- previous generations of cardiac assist devices. The ideal tem-
cardial oxygen consumption. Improved systemic hemodynam- porary LV support system could be quickly and easily inserted
ics including increased cardiac index/cardiac output, in the catheterization laboratory, would offer near complete
decreased pulmonary capillary wedge pressure, and increased cardiac support, and could remain in place for days to weeks.
mean arterial pressure, as well as reduced cardiac work have
also been demonstrated with the Impella LP2.5 (13) and in trials
Table 39.1 Indications for Temporary Cardiac Support
with the TandemHeart (14,15).
l 3-Vessel disease
In addition to improving systemic hemodynamics,
l Acute myocarditis
mechanical LV support has a positive effect on coronary
l Acutely decompensated chronic heart failure
blood flow while reducing myocardial oxygen demand, there-
l Bridge to bridge
fore reducing infarct size. In 1992, our lab demonstrated a l Bridge to destination
modest improvement in regional myocardial blood flow and l Bridge to transplant
concomitant reduced myocardial oxygen demand in a canine l Cardiac arrest
model of anterior MI with LV assistance. When infarction was l Cardiogenic shock
expressed as a percentage of the myocardium at risk, Hemo- l Catheterization laboratory procedures
pump- and IABP-supported animals suffered infarcts of 22% l Heart failure
and 27%, respectively, compared to unsupported controls with l High-risk percutaneous coronary intervention
l High-risk valvular procedures
infarcts of 63%. This difference in infarct size was not signifi-
l Incessant ventricular arrhythmias
cant comparing Hemopump- and IABP-supported animals;
l last remaining patent conduit
however, there was a striking contrast between infarct sizes
l Low ejection fraction with left main disease
in supported versus unsupported animals (16). In the Meyns’ l Mechanical complications of acute myocardial infarction
sheep model of anterior MI, investigators also demonstrated a l Myocardial infarction
significant decrease in the size of infarct in Impella-supported l Papillary muscle rupture
animals. Furthermore, they showed that infarct size was related l Reinitiation of medical therapy
to both timing and the degree of mechanical support. Animals l Ventricular septal defect
supported throughout the infarct and recovery period devel- l Postoperative allograft dysfunction
oped infarctions roughly one-quarter the size of unsupported l Surgical
l Perioperative support of high-risk patients
animals, and approximately half the size of animals receiving
l Failure to wean from bypass
support only during the reperfusion period (12). Adding fur-
l Postcardiotomy syndrome
ther evidence to the theory of ventricular unloading, Remme-
[17/6/010/8:25:34] [504–514]
Traditional ventricular assist devices (VAD) such as the sternal pump, required surgical access. These devices were
MicroMed DeBakey, Jarvic 2000, and the HeartMate II are capable of flows from 3.5 to 5 L/min depending on loading
safe, durable, and capable of complete cardiac assistance, but conditions (26). Later, a 14-Fr device intended for percutaneous
these devices are not well suited for short-term or emergent use femoral insertion and capable of 2.5 L/min was introduced.
(19). A technology that initially showed promise was percuta- Initial testing with Hemopump was promising. In an animal
neous cardiopulmonary bypass (pCPB). model, the Hemopump demonstrated modestly improved
Cardiopulmonary bypass (CPB) was introduced to the regional myocardial blood flow, better LV unloading, and
surgical theater in the early 1950s and the first report of pCPB reduced infarct size when compared to the IABP (16). Initial
was published in 1983 (20). In its simplest form CPB consists of in-human trials also demonstrated improved systemic hemo-
a venous intake cannula, oxygenator, blood pump, and an dynamics and suggested a survival benefit in selected patients
arterial return cannula. In 1990 two independent groups, (11,27). Despite these early successes, further testing demon-
Shawl and colleagues and Vogel et al., reported separate strated an unacceptable incidence of access site complications,
experiences of pCPB-supported coronary angioplasty. Shawl hemolysis, and bleeding (28). The Hemopump did not receive
reported on 121 patients with low EF (<25%) and either an Food and Drug Administration (FDA) approval, but these early
acute coronary syndrome or cardiogenic shock. Vogel’s group experiences laid the groundwork for the next generation of
reported on 105 patients undergoing elective high-risk proce- transvalvular axial blood pumps.
dures—patient EF <25% and/or a target vessel supplying more
than half of the myocardium. Each registry reported a high
degree of procedural success; however, they also encountered INTRA-AORTIC BALLOON PUMP
significant vascular and hematologic complications. Access site With a lack of other easy and quickly implanted devices for
complications ranged from 26% to 46% and the need for transcardiac support, the IABP has been essential in the manage-
fusion was between 30% and 43% (21,22). The frequency of ment of high-risk patients. The IABP was conceived by two
access site complications and bleeding were likely related to the separate groups, Dr Dwight Harken and Drs Adrian and
size of cannulae used, 18 Fr arterial and 20 Fr venous, as well as Arthur Kantrovitz. Dr Harken believed the failing left ventricle
the high level of anticoagulation needed during pump initia- could be unloaded if arterial blood was returned during dias-
tion, activated clotting time (ACT) 400 seconds. Another tole while the aortic valve was closed. The Kantrovitz brothers
limitation of pCPB is the inability to adequately vent the left believed that by achieving peak arterial pressure in diastole,
ventricle. Although pCPB supports systemic circulation, percu- coronary blood flow could be significantly increased. Early
taneous bypass actually increase LV wall stress. In the presence models demonstrated decreased LV end-diastolic pressure, a
of ongoing myocardial ischemia, this increased wall stress leads reduction in LV work index, and increased peak diastolic
to further myocardial damage. In contrast, devices such as the pressure (29). Subsequent developments have resulted in devi-
TandemHeart and Impella actively decompress the LV leading ces that no longer require surgical insertion or large arterioto-
to a beneficial reduction in wall stress. mies, and the current generation of IABP can be placed quickly
Percutaneous CPB-supported PTCA had largely been and easily by experienced cardiologists.
abandoned, but the recent publication of first-in-human trials The IABP is an essential tool for managing unstable AMI
with a new pCPB device, the Lifebridge B2T (Lifebridge patients. Cardiogenic shock generally occurs after at least 40%
Medizintechnik GmbH, Ampfing, Germany), offers tempered of the heart muscle is damaged in the setting of an AMI. A
optimism. The Lifebridge B2T is described as modular “plug- subgroup analysis of the GUSTO-I (Global Utilization of Strep-
and-play” device. Weighing about 20 kg and easily transported, tokinase and Tissue Plasminogen Activator for Occluded Coro-
the device employs 15-Fr arterial and 17-Fr venous access to nary Arteries) trial examined the use of IABC in patients with
circulate up to 4 L of oxygenated blood. Initially demonstrated cardiogenic shock. Of the more than 4000 patients enrolled in
in a swine model, in-human support for three-vessel disease GUSTO-I, 310 underwent IABP placement. Patients receiving
and left main PCI has been promising (23–25). The new device early IABP support had a 30-day mortality rate of 47% compared
does not overcome many of the pitfalls of previous bypass to 60% in patients without device placement. One-year mortality
machines, including suboptimal LV decompression, the large was 57% in the early IABP group and 67% in those patients who
blood-foreign body interface, and limited run time, but the did not receive IABP support (30). The SHOCK trial also explored
intake and outflow cannula are considerably smaller than the question of IABC and AMI. SHOCK (Should We Emergently
predecessors and the modular construction is well suited for Revascularize Occluded Coronaries for Cardiogenic Shock)
emergent use. Further testing is necessary to define a role for investigators examined 1190 patients with cardiogenic shock
the Lifebridge B2T in the modern catheterization laboratory. and AMI. Similar to the GUSTO-I cohort, a subgroup of patients
Surgical VADs are impractical in the acute setting and undergoing IABP placement also showed a significant reduction
pCPB is currently unproven. The TandemHeart offers excellent in mortality with or without revascularization (2).
LV support, but even in skilled hands it may take up to 30 When is hemodynamic support needed for PCI and what
minutes to insert and activate. The balloon pump can be rap- defines “high-risk” PCI? Califf et al. devised the jeopardy
idly inserted; however, an IABP may not adequately support a scoring system based on coronary anatomy and the presence
failing ventricle, particularly in the setting of arrhythmia or of coronary disease as a prognostic indicator for major cardio-
extreme tachycardia—a middle ground is needed. A device vascular outcomes. In 462 non-surgically treated patients,
that can be quickly placed and that provides adequate support investigators demonstrated that patients with a score of 2 had
until definitive therapy can be achieved. The device poised to a 5-year survival of 97%, and those with a score of 12 had a
fill this gap is the catheter-based axial flow pump. 5-year survival of 56% (31). Further analysis indicated that
The first percutaneous axial flow blood pump was devel- LVEF had a similar correlation with prognosis. Adding the
oped and tested by Wampler and colleagues. The Hemopump degree of stenosis to each vessel, in particular the left anterior
(Medtronic Inc.) in its initial forms, a 21-Fr femoral and a 24-Fr descending artery, further increased the prognostic value of the
[17/6/010/8:25:34] [504–514]
jeopardy score. Jeopardy score has been validated and is a good

prognostic indicator in patients undergoing coronary artery
bypass graft surgery and PCI.
Briguori et al. examined the use of prophylactic IABP
placement in 133 high-risk patients undergoing elective PCI
with LV dysfunction (EF <30%) and a jeopardy score >6. Of
these patients, 61 had elective pre-procedural IABP placement
and 72 had conventional PCI. Patients with an IABP suffered
fewer intraprocedural complications (15% vs. 0%; p ¼ 0.001) (32).
Using stepwise regression analysis, high-risk predictors of intra-
procedural complications were provisional versus prophylactic
insertion of an IABP, female sex, and jeopardy score. On the basis
of data from the National Registry of Supported Angioplasty,
older patients (>70 years) and those with left main stenosis are
also at increased risk of intraprocedural complications and may
merit consideration of prophylactic IABP support (22).
The question of when to place an IABP during an AMI
Figure 39.1 Intra-aortic balloon pump in situ. Appropriate position-
has been studied in several trials. PAMI II investigators looked
ing minimizes IABP complications. The distal balloon tip should be
at the use of prophylactic IABP placement in high-risk hemo- between the second and third intercostal spaces and 1 to 2 cm distal
dynamically stable patients with AMI undergoing PTCA. The to the left subclavian artery. Note the balloon marker just distal to the
primary combined endpoint (reduction in death, reinfarction, calcified aortic arch. In this film, the balloon is seen fully inflated in
infarct artery reocclusion, stroke, new-onset heart failure, or the descending aorta. Source: From the Clinical Archives of the
sustained hypotension) showed no significant difference Division of Cardiovascular Medicine, UT Medical School, Houston.
between treatment groups (28.9% vs. 29.2%, p ¼ 0.95). Routine
post-PCI IABP placement was beneficial in terms of reducing
recurrent ischemia and limiting repeat unscheduled catheter- measure until a donor heart becomes available. There is no
izations; however, use of an IABP was also associated with a clinical or statistically significant difference in transplant out-
higher incidence of stroke (2.4% vs. 0%, p ¼ 0.03) (33). In comes between patients who receive IABP support before sur-
another study, Brodie et al. examined prophylactic intra-aortic gery compared to those who do not receive an IABP (39). If
counterpulsation in high-risk AMI. In this series, Brodie com- placed before cardiac transplantation, the IABP should remain
pared placement of an IABP prior to PTCA, after PTCA, and for at least 24 hours postoperatively in the event of allograft
PTCA without IABP support. This study of 1490 AMI patients dysfunction. Potential complications with this method of treat-
found a reduction in catheterization laboratory events—defined ment are increased rates of local and systemic infection and
as ventricular fibrillation, cardiopulmonary arrest, or prolonged prolonged immobilization of the patient while the IABP is in
hypotension—in the group receiving IAPB support prior to place. Left axillary IABP placement may improve patient
PTCA (15% vs. 35%, p ¼ 0.009). IABP support also demonstrated mobility prior to surgery. This method has been studied retro-
benefit in patients with an EF <30% (34). Finally, the SHOCK trial spectively in a small number of patients. In one small series, all
provides critical data regarding the management of AMI and patients with axillary IABP insertion were successfully trans-
shock. The results of the SHOCK trial showed a marked reduc- planted and discharged home (40). The longest time the ambu-
tion in six-month mortality (50% vs. 63%, p ¼ 0.027) in patients latory IABP was in place was for 70 days.
with cardiogenic shock that underwent emergency revasculari- IABP insertion technique is well known to most cardiolo-
zation. In a subgroup analysis, patients that received IABP sup- gists and is only reviewed briefly. Access is obtained via the
port had lower in-hospital mortality than those who did not femoral artery by a modified Seldinger’s technique, and then a
receive an IABP (50% vs. 72%, p ¼ 0.0001) (4). Additionally, long J-tipped wire is advanced to the distal ascending aorta. A
patients that received revascularization and placement of an small incision is made in the skin and bluntly dilated with small
IABP had the lowest in-hospital mortality of all groups analyzed. curved forceps. Once dilatation has been performed, the sheath
Results from these trials strongly support the use of IABC in the and dilator combination are inserted and the dilator is removed.
setting of AMI complicated by cardiogenic shock. The sheath remains in place. The balloon is then advanced to the
Unfortunately, in most of the early trials of the IABC in proximal descending thoracic aorta, distal to the takeoff of the left
the setting of acute STEMI, investigators waited until after subclavian artery (Fig. 39.1). Placement at this level insures there
infarct artery revascularization to initiate support. However, is no obstruction of the renal arteries. Negative suction is applied
as early as 1986 Allen and Buckberg demonstrated the benefi- to the central lumen of the balloon to remove any trapped air and
cial effects of LV decompression on myocardial salvage prior to debris. After flushing the central lumen, the balloon is connected
reperfusion (35). Similarly, in animal models, we have demon- to the pressure tubing and gas line. Once proper placement has
strated a significant improvement in infarct salvage when left been confirmed counterpulsation can begin.
VAD or IABP support was initiated prior to reperfusion com-
pared to support initiated after reperfusion (36,37). The
randomized CRISP-AMI (Counterpulsation Reduces Infarct TANDEMHEART
Size Pre-PCI for Acute Myocardial Infarction) trial will soon Conceptually, the TandemHeart (Cardiac Assist, Inc., Pittsburg,
be conducted to confirm these findings in humans (38). Pennsylvania, U.S.) device is similar to pCPB as described
In patients with severely depressed LV function and earlier; however, key modifications offer substantial benefits
hemodynamic compromise who are currently awaiting cardiac in the setting of AMI and high-risk PCI. Similar to pCPB,
transplantation, an IABP is often placed as a temporizing placement of the TandemHeart requires femoral artery and
[17/6/010/8:25:34] [504–514]
vein access. Additionally, both devices employ an external (ii) an extracorporeal, continuous-flow centrifugal pump,
pump to augment systemic circulation. However, the Tandem- (iii) a 17- to 15-Fr venous return cannula, and (iv) an electronic
Heart uses a unique trans-septal inflow cannula that more pump control module. The inflow cannula has a unique design
effectively unloads the LV during pump operation. The Tan- with a large end hole and multiple side holes permitting high
demHeart also takes advantage of the individual’s own pul- flow rates with limited resistance (Figs. 39.2 and 39.3). The
monary circuit for oxygenation, eliminating the external inflow cannula can be placed via a standard trans-septal
oxygenator and limiting contact between blood elements and approach. We have modified this slightly by employing a soft-
foreign material. Theoretically, this modification minimizes the
inflammatory reaction and allows longer run times.
Dennis et al. were the first to demonstrate the feasibility of
percutaneous left atrial-to-femoral artery bypass. Their initial
design employed a right internal jugular approach and used a
stiff metallic trans-septal cannula (41). After refinements in
materials and methods, the femoral approach was shown in a
small study of animals and critically ill humans (42). Further
modifications lead to the first modern demonstration atrial-
femoral bypass in a cohort of high-risk PCI patients (43). But it
was not until 2001 that Thiele et al. described the initial in-human
trial with the TandemHeart (44). A year after Thiele’s pioneering
work with the TandemHeart for support during cardiogenic
shock, feasibility of TandemHeart-assisted high-risk elective
PCI was described (45,46). Since these initial accounts, multiple
papers have documented successful support of patients with
cardiogenic shock and in patients undergoing high-risk proce-
dures. Most of these accounts are small, descriptive case series
indicating excellent procedural success and relatively limited
complication rates (47–50). Although the aforementioned trials
are noteworthy, two small randomized trials of the IABP versus
the TandemHeart merit further consideration (44,48).
Figure 39.2 Schematic TandemHeart trans-septal inflow cannula.
In the first of these trials, Thiele et al. randomized 41 Multiple side holes and a large end hole permit easy aspiration of
patients suffering from cardiogenic shock following an AMI to oxygenated blood from the left atrium. The 21-Fr inflow catheter
either IABC or TandemHeart support. All patients underwent traverses the interatrial septum at the level of the fossa ovalis. The
revascularization within 24 hours of the onset of cardiogenic catheter then descends the inferior vena cava. Source: Courtesy of
shock. During therapy (mean duration of support 3.5 days for Cardiac Assist, Inc.
the IABP and 4.0 days for the TandemHeart) investigators
closely monitored key hemodynamic parameters and adverse
events. The primary outcome measure, cardiac power index,
was improved more effectively in the TandemHeart group.
Similarly, other common hemodynamic parameters and meta-
bolic indicators of perfusion demonstrated greater improve-
ment in TandemHeart patients. Despite the impressive
hemodynamic differences, investigators were unable to show
a survival benefit. It is likely that the hemodynamic benefits
were offset by an increase in device-related complications
including bleeding, need for transfusion, and limb ischemia.
Investigators also noted higher rates of systemic inflammatory
response syndrome and disseminated intravascular coagula-
tion in the TandemHeart group (44). In a similar study, Burkh-
off and colleagues randomized 19 patients with cardiogenic
shock to TandemHeart support and 14 to IABC. They also
observed improved hemodynamics (decreased pulmonary cap-
illary wedge pressure, increased mean arterial pressure, and
improved cardiac index) with the TandemHeart as compared
to the IABP, but failed to find a survival advantage. The
incidence of many prespecified adverse events was similar
between groups—hemolysis, thrombocytopenia, renal dysfunc-
tion, and infection. But, higher rates of arrhythmia, bleeding,
and access site complications were noted in the TandemHeart Figure 39.3 TandemHeart trans-septal cannulation kit. (Left ) two-
group (48). Because of slow enrolment the trial was ended stage 17-Fr/21-Fr atrial-septal dilator; (center) 13-Fr obturator for
early, with a 30-day survival of 53% (10/19) in the Tandem- aspiration catheter; (right ) 21-Fr aspiration catheter with multiple
Heart group compared to 64% (9/14) in the IABP group. side holes and measuring gradations. Source: Courtesy of Cardiac
In its current design, the TandemHeart device consists of Assist, Inc.
four major components: (i) a 21-Fr trans-septal inflow cannula,
[17/6/010/8:25:34] [504–514]
tipped stiff-core 0.014 in. wire (Sparta Core, Abbot Vascular,

Santa Clara, California) inserted through the trans-septal nee-
dle and directed into the left upper pulmonary vein (LUPV).
Holding the needle and wire fixed, the dilator and trans-septal
sheath are advanced across the septum and into the LUPV. The
wire, needle, and dilator are then removed, and the sheath is
aspirated multiple times to eliminate air and debris. After
placement is confirmed fluoroscopically, a larger, stiff 0.035
in. wire (Amplatz Super Stiff, Boston Scientific, Natick, MA) is
advanced to the LUPV and serves as the working wire. The
interatrial septum is then serially dilated and the aspiration
cannula secured in place. During trans-septal puncture, the
Spartacore wire protects internal structures from inadvertent
laceration with the trans-septal needle. So far, this technique
has proven safe and effective. The external pump unit has a
dual chamber design. The upper chamber houses a low-speed
centrifugal impeller that is in direct contact with blood ele-
ments. The lower chamber is the drive unit consisting of an
electromagnetic pump and liquid bearing. The liquid bearing is
a heparinized solution that functions as local anticoagulation,
pump lubrication, and heat sink. Systemic anticoagulation is
typically withheld until successful trans-septal puncture; how-
ever, during pump assembly and priming we use a target ACT
of 400 seconds. During pump operation an ACT of 200 to
250 seconds is generally sufficient to limit most thrombotic
complications. The control module has a simple operational
design with a fully redundant battery back-up system. Similar
to the Impella device, actual flow rates are not measured.
Instead, pump flow is estimated based on pump speed and
hemodynamic loading conditions. Total cardiac output can be
measured using a thermodilution or continuous monitoring
pulmonary artery catheter. Hemodynamic parameters such as Figure 39.4 Schematic TandemHeart with thigh mounting bracket.
Note same trans-septal cannulation as in Figure 39.2. Also, the
systemic vascular resistance and pulmonary vascular resistance
inflow cannula exits via the right femoral vein. Blood enters the
should be calculated using the total cardiac output measured TandemHeart pump via the central port and is propelled centrifu-
from the pulmonary artery catheter. Figure 39.4 shows the gally out of the side port. Blood returns to the systemic circulation at
TandemHeart fully connected and secured to the thigh- the level of the right iliac via the right femoral artery. Source:
mounting bracket. Figure 39.5 shows an anterior-posterior pro- Courtesy of Cardiac Assist, Inc.
jection of the TandemHeart in situ.
Since introduction of the TandemHeart pVAD in 2001,
the device has been successfully utilized in a multitude of
settings. According to company resources, nearly 1500 proce-
dures have been performed worldwide as of November 2008.
In the past seven years, the TandemHeart has been used for TandemHeart; however, IABC may not be sufficient without
support of cardiogenic shock following AMI and for mechan- some essential level of cardiac output. Bridging this gap is a
ical complications of AMI. In the setting of chronic heart failure, relatively new device, the Impella LP2.5 (Fig. 39.6). The Impella
the TandemHeart has been used as temporary support, i.e., as a family of devices has their basis in the early clinical experiences
bridge to transplant and as a bridge to bridge. In the surgical of the Hemopump (52). Like the Hemopump, the Impella LP2.5
arena, the TandemHeart is effective support for postcardiotomy is a catheter-based transvalvular axial flow pump. The LP2.5 is
syndrome. More recently, case reports have demonstrated benefit the smallest of four devices developed by Abiomed Inc. for
in the setting of critical aortic stenosis, for support during high- temporary cardiac support. The larger devices in the series are
risk mitral valvuloplasty and as effective right VAD (51). Versa- the LD, RD, and the LP5.0. These devices are capable of higher
tility of the TandemHeart has also been demonstrated with flow rates, up to 4 L/min, but require either surgical access or
alternative anitcoagulants, including argatroban and bivalirudin. arterial cutdown for placement. In a randomized trial of 199
In short, the TandemHeart shows promise as a truly percuta- CABG patients, the combination of RD and LD support com-
neous, fully supportive VAD. Although a mortality benefit has pared favorably with normothermic CPB in terms of mortality,
not been demonstrated, increasing patient comorbidities and the perioperative MI, and length of stay (53). These devices have
escalating demand for innovative percutaneous therapies will also demonstrated utility in the setting of shock following AMI,
likely necessitate a greater role for TandemHeart in coming years. acute myocarditis, postcardiotomy syndrome, and as a bridge
to transplant (54–57). The Impella LP2.5 has a 12-Fr maximum
outer diameter and is recommended for placement via a 13- to
IMPELLA 14-Fr sheath. The drive line is 9 Fr and the assembly is mounted
Not all patients presenting with LV compromise or undergoing on a 6-Fr pigtail catheter (Fig. 39.7). The device is advanced
high-risk PCI will require the near complete support of the across the aortic valve over a stiff soft-tipped 0.018 in. coronary
[17/6/010/8:25:34] [504–514]
Figure 39.7 Impella LP2.5. Source: Courtesy of Abiomed Inc.
Figure 39.5 Trans-septal cannulation with the TandemHeart

device. The TandemHeart inflow cannula has been correctly placed
via the trans-septal approach in a patient with prior CAB. Note the
Swan-Ganz catheter in wedge position and surgical clips from the
prior bypass. Severe calcification of the native vessels and previous
stents can also be appreciated. Source: From the Clinical Archives
of the Division of Cardiovascular Medicine, UT Medical School,
Houston.
Figure 39.8 Fluoroscopic RAO view Impella LP2.5. (Upper left)

The opaque drive unit is superior to the aortic valve. (Center ) The
transvalvular cannula traverses aortic valve with the preformed
angle reflecting the normal contour of the LV outflow tract. (Lower
right ) The inflow port and pigtail catheter are in the LV apex.
Abbreviation: RAO, right anterior oblique. Source: Courtesy of
Abiomed Inc.
guidewire (Platinum Plus, Boston Scientific, Natick, Massachu-

setts, U.S.). The Impella LP2.5 is angled to approximate the
couture of the aortic outflow tract. The motor unit and impeller
are located at the proximal end of the device. The intake
cannula sits across the aortic valve with the inflow port near
the cardiac apex (Fig. 39.8). The drive line and infusate port
connect to a control module and purge pump, respectively. The
control module has a simple operational design. Infusate is
maintained under constant pressure, minimum recommended
Figure 39.6 Impella LP2.5 heart schematic. The drive line ascends 300 psi, creating a liquid seal for the motor unit. Flow and aortic
the abdominal aorta. The motor unit, outflow port, and the orifice of pressure are continuously displayed on the control module.
the pressure lumen are positioned appropriately in the proximal Flow reflects an estimated flow rate based on impeller speed
aorta. The flow cannula traverses the aortic valve. The inflow port and transvalvular loading conditions.
and pigtail catheter are at the midcavity level and near the LV apex, Although the Impella LP2.5 lacks the power of its larger
respectively. Source: Courtesy of Abiomed Inc. siblings, the LD and LP5.0, speed and ease of placement make it
an attractive option during elective high-risk procedures and in
[17/6/010/8:25:34] [504–514]
the setting of AMI. Valgimigli and colleagues published the Impella-supported patients. One case of limb ischemia
first in-human use of the Impella LP2.5. The patient, a 56-year- occurred in an Impella patient and transfusion requirements
old man with a history of multiple MIs and reduced EF (27%), were greater in the Impella group; however, mortality was
underwent PCI of a dominant left circumflex artery. Swan- similar in the two groups (46%) (63).
Ganz monitoring during the case demonstrated a marked As of early 2008, Impella devices had been used in nearly
reduction in pulmonary capillary wedge pressure, 18 to 1200 cases worldwide, including 600 Impella LP2.5 procedures.
11 mmHg, during pump operation and a similar increase in The most common application for the LP2.5 has been support
cardiac output, 6.0 L/min compared to 7.4 L/min, as measured during elective high-risk PCI, followed by assistance for car-
by thermodilution (13). Shortly after this initial account, safety diogenic shock, and recovery following AMI. Results reviewed
and feasibility of Impella LP2.5 support were further demon- here have been promising, demonstrating improved hemody-
strated in 19 elective, high-risk PCI patients. All patients had EF namic parameters and excellent patient tolerability. Results of
<40%, and many were older (>60 years), diabetic (53%), and the large ongoing RECOVER II (A Prospective Randomized
had a history of MI (74%). The Impella LP2.5 was successfully Trial Investigating the Use of the IMPELLA RECOVER LP 2.5
placed in all 19 patients and no device-related deaths were System in Patients With Acute Myocardial Infarction Induced
recorded. Two in-hospital mortalities were noted; one follow- Hemodynamic Instability) and PROTECT II trials will likely
ing prolonged cardiac surgery and another secondary to multi- play a significant role in the future of the Impella LP2.5.
organ failure. One patient developed a large access site
hematoma, and two patients required postprocedural blood
transfusion (58). At ACC 2007, Jose Henriques, MD presented a CONTRAINDICATIONS
large series of Impella LP2.5–assisted PCI cases, 109 patients. In As detailed earlier, many of the contraindications to temporary
this series, he found a relatively low rate of in-hospital major LV support are common to the IABP, the TandemHeart, and
adverse cardiac events of 13% (14/109) and excellent tolerabil- the Impella LP2.5. For review, shared contraindications include
ity with 100% device placement (109/109) (59). Abiomed Inc. is severe peripheral vascular disease, moderate to severe aortic
currently enrolling patients in a randomized multicenter trial of valve regurgitation, and uncontrolled bleeding diathesis. In the
high-risk PCI, Impella LP2.5 support versus IABC. This study, case of peripheral arterial disease, vessel diameter is the limit-
PROTECT-II (A Prospective, Multicenter, Randomized Con- ing factor. If the operator is skilled in peripheral interventions,
trolled Trial of the IMPELLA RECOVER LP 2.5 System Versus these limitations may be partially overcome. We have had
Intra Aortic Balloon Pump in Patients Undergoing Non Emer- excellent success with balloon dilation and occasional stenting
gent High Risk PCI), will randomize >600 nonemergent of the target artery prior to device insertion. Next, all of these
patients with low EF and unprotected left main disease, last devices can exacerbate aortic insufficiency; therefore moderate
patent conduit, or three-vessel disease to PCI supported with to severe aortic valve regurgitation is another shared limitation.
either prophylactic IABC or Impella LP2.5 placement. Primary Finally, because of the inherent throbogenicity of foreign mate-
safety endpoints will be death, MI, target vessel revasculariza- rial and the need for adequate anticoagulation during device
tion, and stroke/TIAs. The primary efficacy endpoint will be insertion and operation, uncontrolled bleeding diathesis is a
cardiac power output measured during active support. shared limitation of these temporary LV support devices.
The Impella may also prove to be a valuable tool in the A couple of noteworthy limitations are specific to the
fight against devastating heart attacks; however, experience Impella LP2.5 and the IABP. Because the Impella is a trans-
with the LP2.5 in the acute setting is currently limited. As valvular blood pump, passage of the device across an already
previously discussed, the LP2.5 improves coronary artery stenotic aortic valve may increase the transvalvular gradient
blood flow (17), and mechanical support prior to reperfusion leading to further clinical decompensation. Therefore, insertion
limits infarct size (12,16,35,37,47). In addition, work from in patients with moderate to severe aortic stenosis is not
Tayara et al. suggests that an aggressive approach to AMI recommended. Similarly, passing the Impella through a
patients presenting with cardiogenic shock, including mechan- mechanical prosthetic valve is likely to disrupt valve function.
ical support and provisional transplant, has a significant sur- Therefore, the Impella should not be used in patients with prior
vival benefit (60). A similar aggressive, multitiered approach is valve replacement. In addition to these limitations, the inflow
being evaluated by the Henriques group at the Academic cannula of the Impella sits at the LV apex; therefore, due to the
Medical Center in Amsterdam (61). In a small pilot trial of 20 risk of embolic phenomena, LV mural thrombus is another
AMI patients, 10 receiving Impella support for 3 days following strict contraindication to Impella support. As detailed earlier in
anterior MI and 10 receiving routine care including IABC if this chapter, significant disease of the thoracic or abdominal
needed, researchers demonstrated improved LV recovery with aorta—atherosclerosis or aneurysm—is a shared contraindica-
the Impella LP2.5. At 3-day and 4-month follow-up change in tion of the Impella and the IABP. Finally, although not a contra-
LVEF was greater in the Impella group than in IABP supported indication, the IABP works best when there is ventricle-pump
patients (62). Notably, groups received similar reperfusion synchrony. Although programming modifications and control
therapy, and the Impella cohort appeared somewhat sicker at features have largely addresses this issue, rapid or chaotic
the time of enrollment–lower baseline EF (28% vs. 40%), higher rhythms decrease the efficacy of IABP support. Table 39.2
peak troponin-T (326 vs. 203), higher peak CK (4494 vs. 3897), summarizes the major contraindications to temporary mechani-
and higher NT-proBNP (1619 or 387). Also weighing in on cal LV support.
the topic of mechanical assistance in AMI, results of the ISAR-
SHOCK (Impella LP2.5 vs. IABP in AMI Cardiogenic
Shock) trial are now available. In this small randomized, pro- CLINICAL ASPECTS
spective study of 26 patients presenting with cardiogenic Once a decision has been made that temporary mechanical
shock within 48 hours of AMI, researchers found improved support is required, the interventionalist must select the most
cardiac power index and enhanced serum lactate clearance in appropriate device. In some hospitals, the choice may be
[17/6/010/8:25:34] [504–514]
Table 39.2 Contraindications to Temporary Mechanical Left (17 Fr) arterial access; however, bilateral 12-Fr access is possible
Ventricular Support at the cost of decreasing maximum pump flow rate from 4 to
3 L/min. In a large, adult male with isolated coronary artery
l Mechanical aortic valvesa
l Moderate to severe aortic regurgitation disease, device options are unlimited; however, in the context
l Moderate to severe aortic valve stenosisa of a petite, elderly female with peripheral vascular disease,
l Mural thrombusa arterial cannulation may be impossible without prior periph-
l Severe atherosclerosis of the thoracic or abdominal aortab eral intervention.
l Severe peripheral vascular disease Finally, closure of vascular access sites, relative cost, and
l Significant thoracic or abdominal aortic aneurysm+ physician preferences play a role in device selection. Simple
l Uncontrolled bleeding diathesis manual pressure is often sufficient to ensure safe and durable
a
Applies to Impella only. hemostasis following IABP removal; however, for larger arte-
b
Applies to IABP and Impella. riotomies, we prefer a preclose technique. Our technique
requires deployment of two preloaded vascular sutures prior
to final dilation of the arteriotomy site. During device support,
the suture ends are left untied, tagged with small hemostats,
limited by device availability. Alternatively, the intervention- and then set aside. Following the procedure, while the large
alist must carefully assess the urgency and degree of support vascular conduits are being removed, gentle pressure is applied
required along with access limitations. Finally, issues of closure superior to the access site as the previously deployed sutures
technique, cost considerations, and physician preference play a are tied and advanced. Similar preclose procedures have been
role. All of the devices described in this chapter will not be described elsewhere with excellent (94%) success (64). Next,
available at every hospital. According to Abiomed Inc., as of mid- cost and reimbursement must be considered. The 2008 unit cost
2008 the Impella LP2.5 was available at 100 U.S. hospitals. The of each device is shown in Table 39.3; however, reimbursement
TandemHeart is available at 100 sites worldwide. The IABP has a often varies by diagnosis-related group (DRG). On the basis of
much longer history than either the TandemHeart or the Impella, DRG, payment for the same device, or more accurately the
and thus the IABP is available in most U.S. medical centers. same procedure, may vary by over $20,000. In general, DRGs
Next, the urgency of the patient’s hemodynamic require- for the Impella and TandemHeart will have higher payouts to
ments, the relative degree of cardiac support needed, and the cover the cost of equipment and the additional technical sup-
duration of therapy play a role in device selection. As stated port required. Lastly, it is a very real element of physician
earlier, the TandemHeart offers up to 4 L/min of cardiac preference and appropriate training. Our general cardiology
support; however, in a critically unstable patient, rapid device fellowship provides excellent exposure to the IABP, and most
insertion and early initiation of support may be a greater of our senior fellows feel very comfortable managing patients
concern. Alternatively, in a patient with unstable angina and on balloon pump support. However, there is a significant
severe LV dysfunction scheduled to undergo PCI of an unpro- learning curve with the TandemHeart and the Impella. Even
tected left main, the time necessary to place the TandemHeart is when these more advanced devices are available, only experi-
likely well spent. Finally, the duration of support must be consid- enced operators should consider them as an option. Table 39.3
ered. The TandemHeart is FDA approved for 6 hours of extrac- summarizes the significant clinical aspects discussed in the
orporeal support; however, cases reports of support up to 30 days preceding text.
are available. Similarly, the Impella LP2.5 is intended for short-
term use, but if weaning becomes difficult prolonged support, up
to 7 days, has been reported. IABP support in our institution is CONCLUSION
generally limited to the 24 to 48 hours following an AMI. How- Percutaneous LV support has finally come into the mainstream
ever, the duration of balloon pump support at transplant institu- as a viable option for interventional cardiologists. The Tandem-
tions is frequently on the order of weeks, and prolonged support Heart LV support is the only device currently available that
in excess of three months is not without precedent. provides essentially full support of the circulation, decompres-
There is a significant trade-off between the size of arterial sion of the left ventricle, and supports the circulation at a level
cannula required for a given device and the maximum achiev- commensurate with survival, which is independent of an
able flow rate. The TandemHeart generally requires large bore organized heart rhythm or LV ejection. Unfortunately, it
Table 39.3 Clinical Aspects of Temporary Left Ventricular Support

IABP Impella LP2.5 TandemHeart Lifebridge B2T
Max support (L/min) 0.5 2.5 4.0 5.2
Max duration of support <90 days 6 hr–7 days 6 hr–30 days 6 hr
Access required 7–8.5 Fr arterial 13–14 Fr arterial 15–17 Fr arterial, 15 Fr arterial,
21 Fr venous 17 Fr venous
Estimated time to support (min) <10 <20 <30 30
Closure required Manual Perclose or manual Double preclose or surgical Preclose
Cost $820 $20,000 $19,500 Unknown
Left ventricular loading + ++ +++ Unknown
Comments Fast Intermediate cardiac Excellent support Experimental
Available support Large bore access
Limited support Relatively expensive Superior unloading
Incomplete unloading New
[17/6/010/8:25:34] [504–514]
requires advanced interventional skills, including trans-septal 8. Seung KB, Park DW, Kim YH, et al. Stents versus coronary-artery
puncture and manipulation of a very large cannula into the left bypass grafting for left main coronary artery disease. N Engl J
atrium via the right femoral vein. Additionally, the arterial Med 2008; 358:1781–1792.
return cannula, while possible to insert percutaneously, 9. Hartzler GO, Rutherford BD, McConahay DR, et al. “High-risk”
percutaneous transluminal coronary angioplasty. Am J Cardiol
requires either the preclose technique for temporary use or
1988; 61:33G–37G.
formal femoral artery repair for removal after a more chronic 10. Ellis SG, Roubin GS, King SB III, et al. In-hospital cardiac mortality
implant. The Impella 2.5 device is potentially capable of pump- after acute closure after coronary angioplasty: analysis of risk
ing at a level of 2.5 L/min, but it is dependent on optimal factors from 8,207 procedures. J Am Coll Cardiol 1988; 11:211–216.
loading conditions to achieve this level of support. Obviously 11. Wampler RK, Frazier OH, Lansing AM, et al. Treatment of
the Impella, by itself, is not capable of supporting the entire cardiogenic shock with the Hemopump left ventricular assist
circulation except for very brief periods of time. It does, how- device. Ann Thorac Surg 1991; 52:506–513.
ever, provide active LV decompression and is not dependent 12. Meyns B, Stolinski J, Leunens V, et al. Left ventricular support by
on synchronization with the cardiac cycle. catheter-mounted axial flow pump reduces infarct size. J Am Coll
The IABP is simple to insert but provides indirect sup- Cardiol 2003; 41:1087–1095.
13. Valgimigli M, Steendijk P, Sianos G, et al. Left ventricular unload-
port of diastolic blood flow and a modest level of LV unload-
ing and concomitant total cardiac output increase by the use of
ing. Nonetheless, animal models and observational studies in percutaneous Impella Recover LP 2.5 assist device during high-
humans suggest that the level of support provided by the risk coronary intervention. Catheter Cardiovasc Interv 2005; 65:
balloon pump is sufficient to provide meaningful LV assistance 263–267.
and is associated with reduction in infarct size when implanted 14. Burkhoff D, O’Neill W, Brunckhorst C, et al. Feasibility study of
prior to reperfusion in the setting of STEMI. the use of the TandemHeart percutaneous ventricular assist
Right-sided support is also feasible with the Tandem- device for treatment of cardiogenic shock. Catheter Cardiovasc
Heart device, utilizing two atrial cannulas, one inserted into the Interv 2006; 68:211–217.
superior vena cava and one inserted via antegrade access from 15. Thiele H, Lauer B, Hambrecht R, et al. Reversal of cardiogenic
the right atrium through the right ventricle to the pulmonary shock by percutaneous left atrial-to-femoral arterial bypass assis-
tance. Circulation 2001; 104:2917–2922.
artery. LV assist in the setting of right-sided cardiogenic shock
16. Smalling RW, Cassidy DB, Barrett R, et al. Improved regional
has not been beneficial; however, now that percutaneous right- myocardial blood flow, left ventricular unloading, and infarct
sided cardiac support is available, further evaluation of this salvage using an axial-flow, transvalvular left ventricular assist
technique is warranted. It is not infrequent after implantation of device. A comparison with intra-aortic balloon counterpulsation
a percutaneous VAD, such as the TandemHeart or Impella, for and reperfusion alone in a canine infarction model. Circulation
it to be necessary to wean from the LVAD to IABP support 1992; 85:1152–1159.
prior to complete cessation of LV support. 17. Remmelink M, Sjauw KD, Henriques JP, et al. Effects of left
It is indeed an exciting time in this area. We anticipate ventricular unloading by Impella recover LP2.5 on coronary
that new percutaneous devices will become available in the hemodynamics. Catheter Cardiovasc Interv 2007; 70:532–537.
relatively near future that will significantly improve our arma- 18. Ferguson JJ III, Cohen M, Freedman RJ Jr., et al. The current
practice of intra-aortic balloon counterpulsation: results from the
mentarium for treating circulatory failure without the necessity
Benchmark Registry. J Am Coll Cardiol 2001; 38:1456–1462.
of requiring major cardiac surgery. As new devices become 19. Song X, Throckmorton AL, Untaroiu A, et al. Axial flow blood
available it will be imperative to compare them to the existing pumps. ASAIO J 2003; 49:355–364.
devices and to delineate the relative merits of each device in 20. Phillips SJ, Ballentine B, Slonine D, et al. Percutaneous initiation of
given clinical situations. cardiopulmonary bypass. Ann Thorac Surg 1983; 36:223–225.
21. Shawl FA, Domanski MJ, Wish MH, et al. Percutaneous cardio-
pulmonary bypass support in the catheterization laboratory: tech-
nique and complications. Am Heart J 1990; 120:195–203.
REFERENCES 22. Vogel RA, Shawl F, Tommaso C, et al. Initial report of the National
1. Goldberg RJ, Samad NA, Yarzebski J, et al. Temporal trends in Registry of Elective Cardiopulmonary Bypass Supported Coro-
cardiogenic shock complicating acute myocardial infarction. N nary Angioplasty. J Am Coll Cardiol 1990; 15:23–29.
Engl J Med 1999; 340:1162–1168. 23. Ferrari M, Poerner TC, Brehm BR, et al. First use of a novel plug-
2. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization in and-play percutaneous circulatory assist device for high-risk cor-
acute myocardial infarction complicated by cardiogenic shock. onary angioplasty. Acute Card Care 2008; 10:111–115.
SHOCK Investigators. Should We Emergently Revascularize 24. Jung C, Schlosser M, Figulla HR, et al. Providing Macro- and
Occluded Coronaries for Cardiogenic Shock. N Engl J Med 1999; Microcirculatory Support with the Lifebridge System During
341:625–634. High-risk PCI in Cardiogenic Shock. Heart Lung Circ 2009; 18:
3. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization 296–298.
and long-term survival in cardiogenic shock complicating acute 25. Mehlhorn U, Brieske M, Fischer UM, et al. LIFEBRIDGE: a porta-
myocardial infarction. JAMA 2006; 295:2511–2515. ble, modular, rapidly available “plug-and-play” mechanical cir-
4. Hochman JS, Sleeper LA, White HD, et al. One-year survival culatory support system. Ann Thorac Surg 2005; 80:1887–1892.
following early revascularization for cardiogenic shock. JAMA 26. Wampler RK, Baker BA, Wright WM. Circulatory support of
2001; 285:190–192. cardiac interventional procedures with the Hemopump cardiac
5. McFalls EO, Baldwin D, Marx D, et al. Temporal changes in assist system. Cardiology 1994; 84:194–201.
function and regional glucose uptake within stunned porcine 27. Smalling RW, Sweeney M, Lachterman B, et al. Transvalvular left
myocardium. J Nucl Med 1996; 37:2006–2010. ventricular assistance in cardiogenic shock secondary to acute
6. Hochman JS. Cardiogenic shock complicating acute myocardial myocardial infarction. Evidence for recovery from near fatal
infarction: expanding the paradigm. Circulation 2003; 107:2998–3002. myocardial stunning. J Am Coll Cardiol 1994; 23:637–644.
7. Hannan EL, Wu C, Walford G, et al. Drug-eluting stents vs. 28. Scholz KH, Dubois-Rande JL, Urban P, et al. Clinical experience
coronary-artery bypass grafting in multivessel coronary disease. with the percutaneous hemopump during high-risk coronary
N Engl J Med 2008; 358:331–341. angioplasty. Am J Cardiol 1998; 82:1107–1110, A6.
[17/6/010/8:25:34] [504–514]
29. Bolooki H. Clinical Application of the Intra-Aortic Balloon Pump. 47. Aragon J, Lee MS, Kar S, et al. Percutaneous left ventricular assist
3rd ed. Armonk, NY: Futura Publishing Company, Inc., 1998. device: "TandemHeart" for high-risk coronary intervention. Cath-
30. An international randomized trial comparing four thrombolytic eter Cardiovasc Interv 2005; 65:346–352.
strategies for acute myocardial infarction. The GUSTO investiga- 48. Burkhoff D, Cohen H, Brunckhorst C, et al. A randomized multi-
tors. N Engl J Med 1993; 329:673–682. center clinical study to evaluate the safety and efficacy of the
31. Califf RM, Phillips HR III, Hindman MC, et al. Prognostic value of a TandemHeart percutaneous ventricular assist device versus con-
coronary artery jeopardy score. J Am Coll Cardiol 1985; 5:1055–1063. ventional therapy with intraaortic balloon pumping for treatment
32. Briguori C, Sarais C, Pagnotta P, et al. Elective versus provisional of cardiogenic shock. Am Heart J 2006; 152:469 e1–e8.
intra-aortic balloon pumping in high-risk percutaneous translu- 49. Giombolini C, Notaristefano S, Santucci S, et al. Percutaneous left
minal coronary angioplasty. Am Heart J 2003; 145:700–707. ventricular assist device, TandemHeart, for high-risk percutane-
33. Stone GW, Marsalese D, Brodie BR, et al. A prospective, random- ous coronary revascularization. A single centre experience. Acute
ized evaluation of prophylactic intraaortic balloon counterpulsa- Card Care 2006; 8:35–40.
tion in high-risk patients with acute myocardial infarction treated 50. Kar B, Adkins LE, Civitello AB, et al. Clinical experience with the
with primary angioplasty. Second Primary Angioplasty in Myo- TandemHeart percutaneous ventricular assist device. Tex Heart
cardial Infarction (PAMI-II) Trial Investigators. J Am Coll Cardiol Inst J 2006; 33:111–115.
1997; 29:1459–1467. 51. Giesler GM, Gomez JS, Letsou G, et al. Initial report of percuta-
34. Brodie BR, Stuckey TD, Hansen C, et al. Intra-aortic balloon neous right ventricular assist for right ventricular shock secondary
counterpulsation before primary percutaneous transluminal cor- to right ventricular infarction. Catheter Cardiovasc Interv 2006;
onary angioplasty reduces catheterization laboratory events in 68:263–266.
high-risk patients with acute myocardial infarction. Am J Cardiol 52. Siess T, Nix C, Menzler F. From a lab type to a product: a retro-
1999; 84:18–23. spective view on Impella’s assist technology. Artif Organs 2001; 25:
35. Allen BS, Okamoto F, Buckberg GD, et al. Reperfusion conditions: 414–421.
critical importance of total ventricular decompression during 53. Meyns B, Autschbach R, Boning A, et al. Coronary artery bypass
regional reperfusion. J Thorac Cardiovasc Surg 1986; 92:605–612. grafting supported with intracardiac microaxial pumps versus
36. Achour H, Boccalandro F, Felli P, et al. Mechanical left ventricular normothermic cardiopulmonary bypass: a prospective random-
unloading prior to reperfusion reduces infarct size in a canine ized trial. Eur J Cardiothorac Surg 2002; 22:112–117.
infarction model. Catheter Cardiovasc Interv 2005; 64:182–192. 54. Garatti A, Colombo T, Russo C, et al. Impella recover 100 micro-
37. LeDoux JF, Tamareille S, Felli PR, et al. Left ventricular unloading axial left ventricular assist device: the Niguarda experience. Trans-
with intra-aortic counter pulsation prior to reperfusion reduces plant Proc 2004; 36:623–626.
myocardial release of endothelin-1 and decreases infarction size in 55. Garatti A, Colombo T, Russo C, et al. Different applications for left
a porcine ischemia-reperfusion model. Catheter Cardiovasc Interv ventricular mechanical support with the Impella Recover 100
2008; 72:513–521. microaxial blood pump. J Heart Lung Transplant 2005; 24:481–485.
38. Ohman EM, Patel M. A Multi-center, Randomized, Controlled 56. Garatti A, Colombo T, Russo C, et al. Left ventricular mechanical
Study of Mechanical Left Ventricular Unloading with Counter- support with the Impella Recover left direct microaxial blood
pulsation to Reduce Infarct Size Pre-PCI for Acute Myocardial pump: a single-center experience. Artif Organs 2006; 30:523–528.
Infarction—CRISP AMI. Study Protocol 01. Unknown 2008. 57. Meyns B, Dens J, Sergeant P, et al. Initial experiences with the
39. Christenson JT, Simonet F, Badel P, et al. Optimal timing of Impella device in patients with cardiogenic shock—Impella support
preoperative intraaortic balloon pump support in high-risk coro- for cardiogenic shock. Thorac Cardiovasc Surg 2003; 51:312–317.
nary patients. Ann Thorac Surg 1999; 68:934–939. 58. Henriques JP, Remmelink M, Baan J Jr., et al. Safety and feasibility
40. Cochran RP, Starkey TD, Panos AL, et al. Ambulatory intraaortic of elective high-risk percutaneous coronary intervention proce-
balloon pump use as bridge to heart transplant. Ann Thorac Surg dures with left ventricular support of the Impella Recover LP 2.5.
2002; 74:746–751; discussion 51–52. Am J Cardiol 2006; 97:990–992.
41. Dennis C, Hall DP, Moreno JR, et al. Left atrial cannulation 59. Henriques J. Impella 2.5 Safe to Use in High-Risk PCI: 109 Patient
without thoracotomy for total left heart bypass. Acta Chir Scand Study. ACC, 2007.
1962; 123:267–279. 60. Tayara W, Starling RC, Yamani MH, et al. Improved survival after
42. Glassman E, Engelman RM, Boyd AD, et al. A method of closed- acute myocardial infarction complicated by cardiogenic shock
chest cannulation of the left atrium for left atrial-femoral artery with circulatory support and transplantation: comparing aggres-
bypass. J Thorac Cardiovasc Surg 1975; 69:283–290. sive intervention with conservative treatment. J Heart Lung Trans-
43. Glassman E, Chinitz LA, Levite HA, et al. Percutaneous left atrial plant 2006; 25:504–509.
to femoral arterial bypass pumping for circulatory support in 61. Henriques JP, de Mol BA. New percutaneous mechanical left
high-risk coronary angioplasty. Cathet Cardiovasc Diagn 1993; 29: ventricular support for acute MI: the AMC MACH program.
210–216. Nat Clin Pract Cardiovasc Med 2008; 5:62–63.
44. Thiele H, Sick P, Boudriot E, et al. Randomized comparison of 62. Sjauw KD, Remmelink M, Baan J Jr., et al. Left ventricular
intra-aortic balloon support with a percutaneous left ventricular unloading in acute ST-segment elevation myocardial infarction
assist device in patients with revascularized acute myocardial patients is safe and feasible and provides acute and sustained left
infarction complicated by cardiogenic shock. Eur Heart J 2005; ventricular recovery. J Am Coll Cardiol 2008; 51:1044–1046.
26:1276–1283. 63. Seyfarth M, Sibbing D, Bauer I, et al. A randomized clinical trial to
45. Lemos PA, Cummins P, Lee CH, et al. Usefulness of percutaneous evaluate the safety and efficacy of a percutaneous left ventricular
left ventricular assistance to support high-risk percutaneous cor- assist device versus intra-aortic balloon pumping for treatment of
onary interventions. Am J Cardiol 2003; 91:479–481. cardiogenic shock caused by myocardial infarction. J Am Coll
46. Vranckx P, Foley DP, de Feijter PJ, et al. Clinical introduction of Cardiol 2008; 52:1584–1588.
the Tandemheart, a percutaneous left ventricular assist device, for 64. Lee WA, Brown MP, Nelson PR, et al. Total percutaneous access
circulatory support during high-risk percutaneous coronary inter- for endovascular aortic aneurysm repair ("Preclose" technique). J
vention. Int J Cardiovasc Intervent 2003; 5:35–39. Vasc Surg 2007; 45:1095–1101.
[18/6/010/20:23:13] [515–525]
40
Intracardiac echocardiography
Dennis W. den Uijl, Laurens F. Tops, Nico R. van de Veire, and Jeroen J. Bax
INTRODUCTION St. Jude Medical, West Berlin, New Jersey, U.S.) that generates a
In the past years intracardiac echocardiography (ICE) has 908 wedge-shaped 2-D scanning plane, oriented parallel to the
emerged as a valuable imaging tool for interventional and catheter shaft. The high resolution transducer can be used at
electrophysiological procedures. ICE allows real-time visual- multiple frequencies (5–10 MHz) thereby allowing depth con-
ization of important anatomical structures that cannot be trol and enhancement of tissue penetration to a maximum of
visualized on fluoroscopy, and is not associated with radiation 15 cm. The flexible catheter can be rotated around its axis, and
exposure to the patient and operator. This imaging modality is the tip of the catheter can be deflected by manipulating the
used to guide and monitor interventional procedures and for steering mechanism at the handle of the catheter. The flexibility
early detection of complications. Importantly, as an alternative and maneuverability of the catheter enable the operator to
to transesophageal echocardiography (TEE), ICE can be per- position it inside the right ventricle or coronary sinus. In this
formed without general anesthesia. In this chapter, the basic way, additional views can be obtained, that cannot be acquired
principles and clinical applications of ICE are discussed. from the right atrium. The phased-array catheter is connected
to a dedicated ultrasound machine (SequoiaTM, CypressTM,
CV70TM, Siemens Medical Solutions; and ViewMate1,
ANATOMIC CONSIDERATIONS St. Jude Medical).
ICE is generally performed from within the right atrium or the Several important differences between the two ICE tech-
right ventricle. The images acquired with ICE should therefore nologies exist. The phased-array catheter allows adjustment of
be interpreted from that perspective. Depending on the posi- the ultrasound frequency, thereby enabling depth control. Fur-
tion and direction of the ultrasound catheter, all large cardiac thermore, the phased-array catheter has Doppler capabilities,
structures, including the atria, ventricles, atrioventricular and allowing measurement of hemodynamic and physiological
semilunar valves, coronary sinus, and pericardium can be variables, and has superior flexibility compared with the rota-
visualized with ICE. Awareness of the orientation of the scan- tional catheter. The advantages of rotational ICE include a 3608
ning plane and a good understanding of the three-dimensional scanning plane instead of 908 and the considerably lower costs.
(3-D) cardiac anatomy are essential for a correct interpretation. In daily clinical practice, phased-array ICE is the most com-
monly used technology in the cardiac catheterization labora-
tory. The focus of the present chapter is on phased-array ICE;
EQUIPMENT mechanical intravascular ultrasound is reviewed in chapter 28.
Currently, two different ICE technologies are available. The
first approach utilizes a mechanical ultrasound tipped catheter
which can also be used for endovascular echocardiography. FUNDAMENTALS
The second uses an electronic ultrasound catheter that is ICE can be used to visualize nearly all cardiac structures (1).
equipped with a phased-array transducer at its tip. Both are However, unlike for transthoracic echocardiography (TTE) and
introduced using a retrograde femoral venous approach after TEE, there are no widely accepted standard views for ICE. In
the application of local anesthesia. The mechanical or rotational the following paragraphs, a clinically oriented guide for cath-
system uses a 9-French catheter equipped with a 9-MHz single- eter manipulation and visualization of the various cardiac
element transducer incorporated at its tip (Ultra ICETM, Boston structures is provided.
Scientific, San Jose, California, U.S.). A piezoelectric crystal ICE is generally performed under conscious sedation.
inside the transducer is rotated at 1800 rpm in the radial Using local anesthesia and a femoral vein approach, the ultra-
dimension and provides a two-dimensional (2-D) 3608 scanning sound catheter is inserted through the inferior vena cava into
plane, oriented perpendicular to the catheter shaft. To adjust the right atrium. While standing at the right side of the patient,
the imaging plane, the catheter can be withdrawn or advanced the operator can change the orientation of the ultrasound beam
inside the heart. To prepare the catheter for use, the system has by advancing or withdrawing the catheter and by rotating it
to be filled with 3 to 5 mL sterile water and connected to a around its axis. In this chapter, rotation of the ultrasound
dedicated ultrasound machine (iLabTM System, Galaxy1 Sys- catheter away from the operator is called clockwise rotation,
tem, Galaxy2TM system, or Clearview1 UltraTM system, Boston whereas rotation toward the operator is referred to as counter-
Scientific). clockwise rotation. The orientation of the ultrasound beam can
The electronic system uses an 8-, 9-, or 10-French ultra- also be altered by deflecting the tip of the ultrasound catheter in
sound catheter equipped with a 64-element phased-array trans- two orthogonal planes (anterior-posterior, left-right) through
ducer located at its tip (ACUSON AcunavTM, Siemens Medical manipulation of the two steering knobs at the handle of the
Solutions, Mountain View, California, U.S.; and ViewFlexTM, catheter.
[18/6/010/20:23:13] [515–525]
Even though images acquired with ICE are quite similar

to TEE, the large freedom of ultrasound beam orientation can
easily cause a sense of disorientation to the inexperienced
operator. To gain or regain orientation, a “home view” position
is defined and can be used as the starting point for all catheter
manipulations. In this chapter, home view position is used as
the starting point from which all catheter manipulations are
described. To reach this position, the ultrasound catheter is
positioned in the midright atrium with the control knobs in a
neutral position. The resulting image shows the right atrium,
tricuspid valve, and right ventricle (Fig. 40.1A). The operator
can use these and other anatomical landmarks to maintain
orientation during catheter manipulation. To further improve
the operator’s orientation, a marker is present on the ultra-
sound screen corresponding to the inferior portion of the
ultrasound beam.
Right-Sided Structures
Starting from home view position and slightly withdrawing the
catheter into the inferior right atrium, the Eustachian ridge can
be visualized (Fig. 40.1B). The tissue between the Eustachian
ridge and the tricuspid valve is known as the cavotricuspid
isthmus and is targeted during catheter ablation for atrial
flutter. By advancing the catheter back into home view position
and rotating it counter-clockwise the crista terminalis and right
atrial appendage are visualized. Clockwise rotation will first
bring back home view and will then reveal the right ventricular
outflow tract, the pulmonary artery, and the ascending aorta
(Fig. 40.1B).
Left-Sided Structures
Clockwise rotation of the ultrasound catheter from home view
position, past the right ventricle and right ventricular outflow
tract, provides a view on the left atrium, mitral valve, the
interatrial septum, and the coronary sinus (Fig. 40.1C). By
gently deflecting the catheter tip in the left direction, the left
atrial appendage can be seen (Fig. 40.2A). From this view,
clockwise rotation will reveal the left-sided pulmonary veins
(Fig. 40.2B). The left inferior pulmonary vein is visualized at the
inferior portion of the ultrasound beam and the left superior
pulmonary vein at the superior portion. In case of difficulty
distinguishing between the left superior pulmonary vein and
the left atrial appendage, Doppler flow measurements can be
used to differentiate between the two structures. When further
rotating clockwise, the posterior left atrial wall and the esoph-
agus can be visualized (Fig. 40.2C). Eventually, continued
Figure 40.1 (A) Home view position: RA, RV, TV, and VCI; (B)
clockwise rotation will provide a cross-sectional view of the AoV, cavotricuspid isthmus (*), EuV, and RVOT; (C) interatrial
right-sided pulmonary veins and the right pulmonary artery septum, LA, CS, and MV. Abbreviations: RA, right atrium; RV,
(Fig. 40.2D). Similar to the left pulmonary veins, the right right ventricle; TV, tricuspid valve; VCI, vena cava inferior; AoV,
inferior pulmonary vein is visualized at the inferior portion aortic valve; EuV, Eustachian ridge/valve; RVOT, right ventricular
of the ultrasound beam and the right superior pulmonary vein outflow tract; LA, left atrium; CS, coronary sinus; MV, mitral valve.
at the superior portion.
To acquire a long axis view of the left ventricle and mitral
valve from home view, the catheter is withdrawn slightly into
the inferior right atrium and the tip of the catheter is deflected
in the anterior direction. By advancing the catheter through the
tricuspid valve into the right ventricle and rotating the catheter long axis view, a short axis view of the left ventricle can be
clockwise, the interventricular septum and left ventricle appear acquired by deflecting the tip of the catheter in the left or right
(Fig. 40.3A). This long axis view can be very useful to detect direction (Fig. 40.3B). Advancing or withdrawing the catheter
pericardial effusion during interventional procedures. From the will result in more apical or basal short axis views (Fig. 40.3C).
[18/6/010/20:23:13] [515–525]
INTRACARDIAC ECHOCARDIOGRAPHY 517
Figure 40.2 (A) LA and LAA; (B) descending aorta (Desc Aorta), LA, LSPV, and LIPV; (C) esophagus and posterior LA wall; (D) PA, RIPV,
RSPV, and transverse sinus. Abbreviations: LA, left atrium; LAA, left atrial appendage; LSPV, left superior pulmonary vein; LIPV, left inferior
pulmonary vein; PA, pulmonary artery; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein.
INDICATIONS AND CLINICAL APPLICATIONS Help Interventional Procedures (ICE-CHIP) study was
ICE can be used in a wide variety of diagnostic and interven- designed to address this issue (9). Preliminary results of the
tional procedures. These procedures are summarized in ICE-CHIP study show that ICE has a similar sensitivity for the
Table 40.1 and are reviewed in the following paragraphs. detection of spontaneous contrast, as compared with TEE (10).
In 100 patients with atrial fibrillation, spontaneous contrast
of the left atrium was seen in 50% on ICE and in 55% on TEE
Detection of Intracardiac Thrombus (p ¼ NS) whereas spontaneous contrast of the left atrial append-
Patients undergoing a left-sided interventional procedure are at age was seen in 24% on ICE and in 34% on TEE (p ¼ NS).
high risk for systemic embolism (2,3). ICE can facilitate a safe However, the results on thrombus detection from the ICE-CHIP
left-sided procedure by excluding intracardiac thrombus inside study are not yet available. Therefore, more studies are needed
the left atrial appendage, left atrium, and left ventricle (4). to determine the exact value of ICE for the detection of intra-
Furthermore, ICE can be used to assess the presence of spon- cardiac thrombi.
taneous contrast, thereby identifying patients at high risk for
thrombus formation (5). ICE can help to detect the formation of
thrombi at an early phase and allow for treatment prior to the Closure of Atrial Septal Defect
occurrence of embolic events (6,7). Percutaneous transcatheter device closure of atrial septal defect
The efficacy of TEE to detect intracardiac thrombus has or patent foramen ovale (PFO) has proven to be a safe and
been established by the Assessment of Cardioversion Using effective alternative to open heart surgery (11,12). While per-
Transesophageal Echocardiography (ACUTE) trial (8). Even cutaneous closure of PFO may be performed under fluoroscopy
though ICE provides high quality images comparable to TEE, guidance only, closing procedures of atrial septal defect are
only a few studies have compared the sensitivity of the two typically guided by TEE and fluoroscopy. However, ICE does
imaging modalities for the detection of intracardiac thrombus. not require general anesthesia, and may provide similar images
The Intracardiac Echocardiography-Guided Cardioversion to as TEE (13,14). It has been shown that the use of ICE during
[18/6/010/20:23:13] [515–525]
Table 40.1 Applications of Intracardiac Echocardiography During

Interventional and Electrophysiological Procedures
l Detection of intracardiac thrombus
l Closure of atrial septal defect
l Transseptal puncture
l Electrophysiological procedures
Atrial fibrillation ablation
Complex atrial flutter ablation
Ventricular tachycardia ablation
Left ventricular lead placement in cardiac resynchronization
therapy
l Other interventional procedures
Biopsy of an intracardiac tumor
Left atrial appendage closure
Closure of ventricular septal defect
Alcohol septum ablation in hypertrophic obstructive cardio-
myopathy
Mitral valve balloon valvuloplasty
Percutaneous valve procedures
To guide the placement of a transcatheter closure device,

the ultrasound catheter is positioned in the home view position
and is rotated clockwise to visualize the interatrial septum and
fossa ovalis (Fig. 40.4A, B, C). By using Doppler capacities, the
flow between the left and right atrium can be visualized
and quantified (Fig. 40.4D). A guiding wire is then placed
through the atrial septal defect and inside the left atrium.
Subsequently, the catheter that contains the closure device is
advanced through the atrial septal defect and the left-sided
portion of the occluder is deployed. After this step, the position
of the device against the interatrial septum is carefully eval-
uated before deploying the right-sided portion of the occluder
to avoid malposition and the associated risk of migration of the
device (Fig. 40.5A). Once the operator is convinced that the
position is correct, the right-sided portion of the occluder is
deployed (Fig. 40.5B). Once again the position and the stability
of the device are checked and them the occluder is released.
Transseptal Puncture
A transseptal puncture provides antegrade access to the left
atrium and left ventricle during left-sided interventional proce-
dures as an alternative to a retrograde approach through the
aortic valve and mitral valve. However, a transseptal puncture
can result in serious complications, such as aortic perforation,
pericardial tamponade, and perforation of the inferior vena cava
(17). The fossa ovalis is considered to be the safest site to
Figure 40.3 (A) Long axis view of the LV, IVS, and MV; (B) short
perform a transseptal puncture to avoid these complications.
axis view at the apical level; (C) short axis view at the basal level.
Abbreviations: LV, left ventricle; IVS, interventricular septum; MV,
ICE allows excellent visualization of the fossa ovalis and can be
mitral valve. used to detect a PFO or monitor the transseptal puncture (18). At
present, no prospective studies have addressed the question of
whether ICE may improve the safety of transseptal punctures.
To visualize the interatrial septum, the catheter is gently
rotated clockwise from home view position. The interatrial
transcatheter device closure may result in a reduction of fluo- septum consists of a thicker part (limbus) and thin part (fossa
roscopy time (9.5 1.6 vs. 6.0 1.7 minutes, p < 0.0001) (13), ovalis) (Fig. 40.6A). To detect a PFO, saline/contrast is injected
procedure length (47 8 vs. 35 6 minutes, p < 0.001), and through the femoral vein inside the right atrium and the patient
catheterization laboratory occupation (92 18 vs. 50 12 is instructed to perform the Valsalva maneuver (Fig. 40.7A). In
minutes, p < 0.001) compared with TEE guided interventions the presence of a patent foramen the contrast will cross the
(15). Importantly, the high costs of an ICE catheter may be interatrial septum, into the left atrium (Fig. 40.7B). In the
balanced by the need for general anesthesia during TEE guided absence of a PFO, a transseptal sheath with a concealed Brock-
procedures (16). enbrough transseptal needle is inserted through the femoral
[18/6/010/20:23:13] [515–525]
Figure 40.4 (See color insert, only D) (A) IAS, LA, and RA. (B) During a Valsalva maneuver, the PFO is revealed. (C) A large type II atrial
septal defect (two markers). (D) Doppler flow delineates the flow across PFO during a Valsalva maneuver. Abbreviations: IAS, interatrial
septum; LA, left atrium; RA, right atrium; PFO, patent foramen ovale.
Figure 40.5 (A) A biodegradable closure device is inserted across the IAS inside LA. Subsequently, the left-sided occluder is deployed.
(B) After confirmation of the position of the device, the right-sided occluder is also deployed. Abbreviations: IAS, interatrial septum; LA, left
atrium; RA, right atrium.
[18/6/010/20:23:13] [515–525]
Figure 40.6 (A) Fossa ovalis, LA, and LIPV; (B) tenting of the transseptal sheath against the fossa ovalis. Abbreviations: LA, left atrium; RA,
right atrium; LIPV, left inferior pulmonary vein.
Figure 40.7 (A) Contrast inside the RA during Valsalva in a patient with a closed foramen ovale. (B) Contrast crosses the IAS from the RA to
the LA during Valsalva in a patient with a patent foramen ovale. Abbreviations: IAS, interatrial septum; LA, left atrium; RA, right atrium.
vein inside the right atrium. Using fluoroscopy and ICE, the myocardium. Moreover, ICE can be used to visualize morpho-
transseptal sheath is positioned against the fossa ovalis. In case logical changes in the myocardium, such as increased echo
of a stable position of the sheath against the fossa ovalis, a density, wall thickening and crater formation as a sign of
“tenting” phenomenon can be seen on ICE (Fig. 40.6B). The effective lesion formation (22), and the development of micro
transseptal puncture can now be performed by pushing the bubbles as a sign of tissue heating and potential char formation
needle out from the sheath, through the fossa ovalis. Successful (23,24). In the following paragraphs, the specific role of ICE in
transseptal puncture can be confirmed on ICE by injecting various electrophysiological procedures is reviewed.
saline/contrast through the needle inside the left atrium.
Atrial Fibrillation Ablation
Electrophysiological Procedures Radiofrequency catheter ablation targeting the pulmonary
ICE has become an important imaging tool during electro- veins is a potential curative treatment option for patients with
physiological procedures. In addition to thrombus detection drug-refractory atrial fibrillation (25,26). However, it is associ-
and guidance of a transseptal puncture, ICE can be used to ated with long procedure times and a small risk for severe
identify key anatomical structures to facilitate complex proce- complications, including pulmonary vein stenosis, systemic
dures like atrial fibrillation ablation and atrial flutter ablation embolism, cardiac tamponade, and esophagus injury (2). ICE
(19–21). ICE can visualize the exact location of the mapping can facilitate these complex procedures by visualization of the
catheter and confirm stable contact of the catheter against the pulmonary veins and monitoring of the location of the ablation
[18/6/010/20:23:13] [515–525]
catheter, and may help in avoiding complications by visualiza- the development of an atrioesophageal fistula (27,33,34). While
tion of the esophagus and other important surrounding struc- monitoring the relation between the esophagus and the ablation
tures (21,23,27). catheter with ICE, the ablation power and duration can be
Several studies have shown that pulmonary vein anat- adjusted to reduce esophageal damage (27). Monitoring lesion
omy is highly variable (28–30). Application of radiofrequency development and the occurrence of micro bubbles as an indica-
current inside a pulmonary vein ostium may cause pulmonary tion of an increased esophageal temperature, allows the opera-
vein stenosis and pulmonary hypertension (31). Accurate visu- tor to perform additional energy titration, thereby further
alization of the pulmonary vein region is therefore of utmost minimizing the risk of esophageal damage (27,32).
importance to safely and effectively perform catheter ablation. As an alternative to anatomical guidance with ICE, image
A head-to-head comparison between ICE and multislice com- integration with MSCT or magnetic resonance imaging (MRI)
puted tomography (MSCT) demonstrated that ICE enables integration is commonly used to guide radiofrequency catheter
accurate assessment of pulmonary vein anatomy and mean ablation for atrial fibrillation (35). A 3-D image of the left atrium
ostial diameters (ICE 1.51 0.22 vs. MSCT 1.45 0.29 mm, p ¼ can be integrated with an electroanatomical map by performing
NS). However, fewer additional pulmonary veins were a semiautomatic registration process. However, the validity of
detected using ICE, as compared with MSCT (ICE 2 ¼ 8% vs. this technique is largely dependent on the quality of the regis-
MSCT 5 ¼ 21%) (30). This was confirmed by Jongbloed et al. tration (36–38). An inaccurate registration process can result in
who detected fewer additional pulmonary veins with ICE, as a large shift of landmark points of up to 5 to 10 mm, thereby
compared with MSCT (ICE 7 ¼ 17% vs. MSCT 13 ¼ 32%) (28). compromising the safety and efficiency of lesion placement
In addition, an underestimation of the ostial diameter on ICE (36,37). Adjunctive real-time imaging with ICE can be used to
compared with the ostial diameter in superior-inferior direction confirm the accuracy of the registration process to ensure an
on MSCT was noted (14.9 4.0 vs. 18.4 3.4 mm, p < 0.01) (28). accurate delivery of radiofrequency energy (36).
This finding suggests the need for 3-D imaging to accurately An electroanatomical mapping system (CARTO-
visualize the shape and dimensions of the pulmonary vein SOUNDTM, Biosense Webster, Diamond Bar, California, U.S.)
ostia. Nevertheless, in a group of 259 patients, Marrouche et al. allows integration of ICE and electroanatomical mapping (39).
demonstrated that anatomical guidance of radiofrequency cath- By integrating ICE and electroanatomical mapping, an accurate
eter ablation with ICE is both safe and effective (23). Impor- 3-D anatomical shell of the left atrium and pulmonary veins can
tantly, the use of ICE in addition to a circular catheter resulted be acquired without performing a registration process (40). A
in an improved outcome compared with a circular catheter modified phased-array ultrasound catheter with an imbedded
alone (23). navigation sensor at its tip (SoundstarTM, Biosense Webster) is
The esophagus and left atrial posterior wall are very positioned inside the right atrium. The mapping system can
closely related. With the use of ICE, Ren and colleagues dem- detect the position and direction of the ICE catheter, thereby
onstrated that the mean distance between the left atrial posterior enabling the projection of the scanning plane inside its 3-D
wall and the esophagus was 5.8 1.2 mm (range 3.2–10.1 mm) environment. By gently rotating the ultrasound catheter, ECG-
and that the left atrial posterior wall and the esophagus were gated images of the left atrium and pulmonary veins are
contiguous over a mean length of 36.0 7.7 mm (range 18–59 acquired. On each image, the endocardial borders (contours)
mm) (27). As a consequence, the temperature inside the esoph- are traced manually and are thereafter assigned to a designated
agus may increase significantly during left atrial ablation (32). map (Fig. 40.8A). Separate maps are created for the left atrial
Heating of the esophagus can result in esophageal injury vary- body and each of the pulmonary veins. All contours within a
ing from transient erythematous changes to tissue necrosis and map are used to create a 3-D shell of the structure (Fig. 40.8B).
Figure 40.8 (See color insert) (A) After acquisition of an ECG-gated ultrasound image, the endocardial contours (green) of the LA and
pulmonary veins are manually traced and are thereafter assigned to a corresponding map. (B) By systematically collecting images of the
whole LA and marking the endocardial borders, a registered 3-D reconstruction of the LA anatomy is created. (C) The acquired 3-D geometry
can be used to guide radiofrequency catheter ablation for atrial fibrillation. Abbreviation: LA, left atrium.
[18/6/010/20:23:13] [515–525]
By combining all maps, the 3-D geometry of the whole left atrium segments as seen on ICE, the ischemic substrate could be
and pulmonary veins is visualized and can be merged with a mapped and the ablation procedure could be performed suc-
MSCT image to facilitate the ablation procedure (Fig. 40.8C). cessfully.
Complex Atrial Flutter Ablation Left Ventricular Lead Placement in Cardiac Resynchronization
A common atrial flutter is an organized tachycardia with a Therapy
reentry circuit inside the right atrium and a protected isthmus Cardiac resynchronization therapy (CRT) has a beneficial effect
between the tricuspid valve and the inferior vena cava (cavo- on clinical symptoms, exercise capacity, and left ventricular
tricuspid isthmus) (20). Ablation of a common flutter is per- systolic function in selected patients with drug-refractory heart
formed by creating a linear line of block across the failure (48–50). Moreover, CRT is associated with an increased
cavotricuspid isthmus (41). Even though it is considered unnec- survival and a reduction in the number of rehospitalizations for
essary to use special imaging or mapping during a standard heart failure, as compared with optimal medical treatment (50).
procedure, during complex cases ICE may be used to facilitate However, implantation of a CRT device—usually performed
the procedure (19). ICE can identify the cavotricuspid isthmus under fluoroscopic and angiographic guidance—can be chal-
and other anatomical structures that can act as electrical bar- lenging because of venous anatomy, resulting in a failure to
riers during atrial flutter, like the crista terminalis and Eusta- place the left ventricular pacing lead in up to 8% of patients
chian ridge (20). Ablation of an atrial flutter can be complicated (48–50). A number of case studies report on the use of ICE to
by complex anatomy, for example, in patients with previous visualize the coronary sinus to guide left ventricular lead
surgery for congenital heart disease. Particularly in these placement (51,52). However, at present no prospective studies
patients, ICE can facilitate the ablation procedure by visualiz- have reported a beneficial effect of ICE guidance on the success
ing important anatomical structures and guiding catheter rate for left ventricular lead placement.
placement (42).
Ventricular Tachycardia Ablation Other Interventional Procedures

Ablation of ventricular tachycardia is usually limited to indu- Biopsy of Intracardiac Mass
cible and tolerated arrhythmias (43,44). Techniques to identify ICE can be used to visualize the origin and extent of an
the arrhythmogenic substrate without inducing the tachycardia intracardiac mass (Fig. 40.9A). Therefore, ICE may be used to
are being developed to treat patients who do not meet these guide biopsies (Fig. 40.9B) and to monitor associated compli-
criteria. ICE allows identification of akinetic and dyskinetic cations such as perforation or bleeding as suggested by few
(aneurysmatic) myocardial segments in patients with ischemic preliminary reports (53,54).
ventricular tachycardia, thereby visualizing the exact location
and extent of the substrate (45,46). Furthermore, ICE can be Left Atrial Appendage Closure
used to visualize small aneurysms of the right ventricle in Implantation of a left atrial appendage occlusion device has
patients with (suspected) arrhythmogenic right ventricular been advocate as strategy to reduce the risk for systemic
dysplasia, thereby detecting the arrhythmogenic substrate in embolism in patients with atrial fibrillation and contraindica-
these patients (46). tions for anticoagulation, and the procedure is typically guided
Recently, the feasibility of the integration of ICE and by TEE. Recently, in a small group of patients the feasibility of
electroanatomical mapping to guide ischemic ventricular tachy- ICE guidance as an alternative to TEE was reported (55). ICE
cardia ablation was demonstrated (47). By creating a 3-D geom- provided similar visualization of the left atrial appendage and
etry of the left ventricle and marking the akinetic and dyskinetic similar assessment of the left atrial appendage orifice diameter
Figure 40.9 (A) An intracardiac mass originating from the superior vena cava is extending into the RA and tricuspid valve. (B) Intracardiac
echocardiography is used to monitor and guide the biopsy by visualizing both tumor and bioptome. Abbreviations: RA, right atrium; RV, right
ventricle.
[18/6/010/20:23:13] [515–525]
compared with TEE (ICE 22.6 3.4 vs. TEE 19.5 1.5 mm, p ¼ setting (63,64). In 10 patients, endocardial structures could be
NS). Importantly, the degree of accuracy with respect to exclu- visualized in great detail from various angles (63). The ability to
sion of a thrombus inside the left atrial appendage, the exact visualize cardiac anatomy from different angles could benefit
positioning of the delivery sheath, and verification of the loca- catheter navigation. However, this invasive approach is limited
tion and stability of the occlusion device using ICE were was to patients undergoing epicardial access for catheter ablation
comparable to TEE. (63,64).
Ventricular Septal Defect Closure

ICE can be used to guide transcatheter closure of a perimem-
LIMITATIONS
branous ventricular septal defect (56). This imaging modality
Although phased-array ICE enables adjustment of the ultra-
allows identification and visualization of the defect and mon-
sound frequency, tissue penetration remains a limiting factor in
itoring the placement of a guiding wire through the defect and
visualizing cardiac anatomy. Use of a lower ultrasound fre-
the deployment of the left-sided occluder. Subsequently, ICE is
quency would result in a higher degree of tissue penetration
used to confirm the correct position of the left-sided occluder
allowing visualization of structures further away from the
against the interventricular septum before the deployment of
transducer, but at the cost of lower image resolution. In addi-
the right-sided occluder. After deployment of the second
tion, the costs of phased-array ICE are relatively high as
occluder, ICE can be used to assess any residual shunt and
compared with TEE and these expensive catheters are for single
valvular regurgitation that may have resulted from the proce-
use only. However, the costs of ICE are somewhat balanced by
dure. In 12 patients, Cao et al. documented that ICE may be
the need for general anesthesia and an echocardiographist
used as a safe and effective alternative for TEE to guide closure
during TEE. Moreover, ICE provides 2-D monoplane images.
of a ventricular septal defect (56).
This limitation can be partially overcome by the flexibility of
the catheter enabling to visualize the same structure from
Alcohol Ablation in Hypertrophic Obstructive Cardiomyopathy
another angle. Nevertheless, operators who are used to multi-
Alcohol septal ablation is an effective treatment to reduce the
plane TEE may still have difficulty obtaining the same views.
intraventricular gradient in patients with hypertrophic obstruc-
Finally, there are no widely accepted standard views for ICE, in
tive cardiomyopathy (57). However, the efficacy and safety of
contrast to TEE and TTE. This may be difficult, in particular for
the procedure is dependent on the identification of the correct
the inexperienced operator. Standard manipulation of the ultra-
septal artery. To identify this branch, echocontrast is commonly
sound catheter starting from home view position as well as
injected into a septal artery at the time of coronary angiography
recognition of landmark structures may be helpful.
and TTE is used to detect the extent and localization of the
corresponding myocardial territory. ICE allows high quality
visualization of the entire interventricular septum and may be a SPECIAL ISSUES
useful tool to guide alcohol septal ablation (58). In nine ICE is an invasive imaging modality and its use is usually
patients, Pedone et al. demonstrated that use of ICE to guide confined to patients undergoing a percutaneous interventional
the procedure was feasible (59). However, more studies are procedure. In general, the contraindications for ICE are similar
needed to define the role of ICE in alcohol septal ablation in to other right-sided cardiac catheterization procedures using a
patients with hypertrophic obstructive cardiomyopathy. transfemoral access. In pediatric patients, the use of ICE is
limited by the respective diameters of the femoral vein and the
Mitral Valve Balloon Valvuloplasty ultrasound catheter.
Percutaneous balloon valvuloplasty is an accepted alternative
to surgical commissurotomy in selected patients with symp-
tomatic mitral stenosis. In this setting, ICE can be used to
CONCLUSIONS
exclude thrombus formation at the level of the left atrium,
ICE is a valuable imaging tool for a wide variety of interven-
assess the morphology and function of the mitral valve, guide
tional and electrophysiological procedures. This imaging
the transseptal puncture, monitor the positioning of the bal-
modality allows real-time visualization of anatomical struc-
loon, and assess any residual valvular gradient or postproce-
tures, catheters, and devices thereby enabling the monitoring
dural mitral regurgitation (60). In addition, it may allow an
and guidance of complex procedures like catheter ablation for
early detection of complications such as cardiac tamponade.
atrial fibrillation and placement of a transcatheter closure
device for atrial or ventricular septal defects. Since it provides
Percutaneous Valve Procedures
images of quality comparable to TEE, ICE may be used—in the
ICE may provide online anatomical information useful to guide
hands of an experienced operator—as an alternative to TEE
percutaneous valve repair or replacement (61,62). Accordingly,
during closure of an atrial septal defect or ventricular septal
this imaging modality may be used to determine the appropri-
defect and during percutaneous occlusion of the left atrial
ate size and site of deployment of the percutaneous valve as
appendage. In addition, ICE is a potentially safe alternative
well as to monitor the anatomical and functional result of the
for TEE to detect an intracardiac thrombus.
procedure (61). Studies are needed to define the role of ICE
during percutaneous valve procedures.
REFERENCES
Intrapericardial Use of ICE 1. Packer DL, Stevens CL, Curley MG, et al. Intracardiac phased-
Positioning the ICE catheter inside the pericardium has the array imaging: methods and initial clinical experience with high
potential to provide valuable information during complex abla- resolution, under blood visualization: initial experience with
tion procedures. Recently, the safety and feasibility of this intracardiac phased-array ultrasound. J Am Coll Cardiol 2002;
approach was demonstrated in both experimental and clinical 39:509–516.
[18/6/010/20:23:13] [515–525]
2. Cappato R, Calkins H, Chen SA, et al. Worldwide survey on the 23. Marrouche NF, Martin DO, Wazni O, et al. Phased-array intra-
methods, efficacy, and safety of catheter ablation for human atrial cardiac echocardiography monitoring during pulmonary vein
fibrillation. Circulation 2005; 111:1100–1105. isolation in patients with atrial fibrillation: impact on outcome
3. Thakur RK, Klein GJ, Yee R, et al. Embolic complications after and complications. Circulation 2003; 107:2710–2716.
radiofrequency catheter ablation. Am J Cardiol 1994; 74:278–279. 24. Wazni OM, Rossillo A, Marrouche NF, et al. Embolic events and
4. Jongbloed MR, Bax JJ, van der Wall EE, et al. Thrombus in the left char formation during pulmonary vein isolation in patients with
atrial appendage detected by intracardiac echocardiography. Int J atrial fibrillation: impact of different anticoagulation regimens and
Cardiovasc Imaging 2004; 20:113–116. importance of intracardiac echo imaging. J Cardiovasc Electro-
5. Ren JF, Marchlinski FE, Callans DJ, et al. Increased intensity of physiol 2005; 16:576–581.
anticoagulation may reduce risk of thrombus during atrial fibril- 25. Haissaguerre M, Jais P, Shah DC, et al. Spontaneous initiation of
lation ablation procedures in patients with spontaneous echo atrial fibrillation by ectopic beats originating in the pulmonary
contrast. J Cardiovasc Electrophysiol 2005; 16:474–477. veins. N Engl J Med 1998; 339:659–666.
6. Keane D, Mansour M, Singh J. Detection by intracardiac echocar- 26. Pappone C, Rosanio S, Oreto G, et al. Circumferential radiofre-
diography of early formation of left atrial thrombus during pul- quency ablation of pulmonary vein ostia: a new anatomic
monary vein isolation. Europace 2004; 6:109–110. approach for curing atrial fibrillation. Circulation 2000; 102:
7. Ren JF, Marchlinski FE, Callans DJ. Left atrial thrombus associated 2619–2628.
with ablation for atrial fibrillation: identification with intracardiac 27. Ren JF, Lin D, Marchlinski FE, et al. Esophageal imaging and
echocardiography. J Am Coll Cardiol 2004; 43:1861–1867. strategies for avoiding injury during left atrial ablation for atrial
8. Klein AL, Grimm RA, Murray RD, et al. Use of transesophageal fibrillation. Heart Rhythm 2006; 3:1156–1161.
echocardiography to guide cardioversion in patients with atrial 28. Jongbloed MR, Bax JJ, Lamb HJ, et al. Multislice computed
fibrillation. N Engl J Med 2001; 344:1411–1420. tomography versus intracardiac echocardiography to evaluate
9. Rao HB, Saksena S, Mitruka R. Intra-cardiac echocardiography the pulmonary veins before radiofrequency catheter ablation of
guided cardioversion to help interventional procedures (ICE- atrial fibrillation: a head-to-head comparison. J Am Coll Cardiol
CHIP) study: study design and methods. J Interv Card Electro- 2005; 45:343–350.
physiol 2005; 13(suppl 1):31–36. 29. Marom EM, Herndon JE, Kim YH, et al. Variations in pulmonary
10. Nagarakanti R, Saksena S, Sra J, et al. Cardiac arrhythmias. J Am venous drainage to the left atrium: implications for radiofre-
Coll Cardiol 2005; 45:A91–A128. quency ablation. Radiology 2004; 230:824–829.
11. Du ZD, Hijazi ZM, Kleinman CS, et al. Comparison between 30. Wood MA, Wittkamp M, Henry D, et al. A comparison of pul-
transcatheter and surgical closure of secundum atrial septal defect monary vein ostial anatomy by computerized tomography, echo-
in children and adults: results of a multicenter nonrandomized cardiography, and venography in patients with atrial fibrillation
trial. J Am Coll Cardiol 2002; 39:1836–1844. having radiofrequency catheter ablation. Am J Cardiol 2004;
12. Jones TK, Latson LA, Zahn E, et al. Results of the U.S. multicenter 93:49–53.
pivotal study of the HELEX septal occluder for percutaneous 31. Robbins IM, Colvin EV, Doyle TP, et al. Pulmonary vein stenosis
closure of secundum atrial septal defects. J Am Coll Cardiol after catheter ablation of atrial fibrillation. Circulation 1998;
2007; 49:2215–2221. 98:1769–1775.
13. Bartel T, Konorza T, Arjumand J, et al. Intracardiac echocardiogra- 32. Cummings JE, Schweikert RA, Saliba WI, et al. Assessment of
phy is superior to conventional monitoring for guiding device temperature, proximity, and course of the esophagus during
closure of interatrial communications. Circulation 2003; 107:795–797. radiofrequency ablation within the left atrium. Circulation 2005;
14. Mullen MJ, Dias BF, Walker F, et al. Intracardiac echocardiogra- 112:459–464.
phy guided device closure of atrial septal defects. J Am Coll 33. Marrouche NF, Guenther J, Segerson NM, et al. Randomized
Cardiol 2003; 41:285–292. comparison between open irrigation technology and intracar-
15. Boccalandro F, Muench A, Salloum J, et al. Interatrial defect sizing diac-echo-guided energy delivery for pulmonary vein antrum
by intracardiac and transesophageal echocardiography compared isolation: procedural parameters, outcomes, and the effect on
with fluoroscopic measurements in patients undergoing percuta- esophageal injury. J Cardiovasc Electrophysiol 2007; 18:583–588.
neous transcatheter closure. Catheter Cardiovasc Interv 2004; 34. Pappone C, Oral H, Santinelli V, et al. Atrio-esophageal fistula as a
62:415–420. complication of percutaneous transcatheter ablation of atrial fibril-
16. Alboliras ET, Hijazi ZM. Comparison of costs of intracardiac lation. Circulation 2004; 109:2724–2726.
echocardiography and transesophageal echocardiography in mon- 35. Tops LF, Bax JJ, Zeppenfeld K, et al. Fusion of multislice com-
itoring percutaneous device closure of atrial septal defect in chil- puted tomography imaging with three-dimensional electroana-
dren and adults. Am J Cardiol 2004; 94:690–692. tomic mapping to guide radiofrequency catheter ablation
17. Lundqvist C, Olsson SB, Varnauskas E. Transseptal left heart procedures. Heart Rhythm 2005; 2:1076–1081.
catheterization: a review of 278 studies. Clin Cardiol 1986; 9:21–26. 36. Daccarett M, Segerson NM, Gunther J, et al. Blinded correlation
18. Epstein LM, Smith T, TenHoff H. Nonfluoroscopic transseptal study of three-dimensional electro-anatomical image integration
catheterization: safety and efficacy of intracardiac echocardio- and phased array intra-cardiac echocardiography for left atrial
graphic guidance. J Cardiovasc Electrophysiol 1998; 9:625–630. mapping. Europace 2007; 9:923–926.
19. Morton JB, Sanders P, Davidson NC, et al. Phased-array intra- 37. Fahmy TS, Mlcochova H, Wazni OM, et al. Intracardiac echo-
cardiac echocardiography for defining cavotricuspid isthmus guided image integration: optimizing strategies for registration. J
anatomy during radiofrequency ablation of typical atrial flutter. Cardiovasc Electrophysiol 2007; 18:276–282.
J Cardiovasc Electrophysiol 2003; 14:591–597. 38. Zhong H, Lacomis JM, Schwartzman D. On the accuracy of
20. Olgin JE, Kalman JM, Fitzpatrick AP, et al. Role of right atrial CartoMerge for guiding posterior left atrial ablation in man.
endocardial structures as barriers to conduction during human Heart Rhythm 2007; 4:595–602.
type I atrial flutter. Activation and entrainment mapping guided 39. Khaykin Y, Klemm O, Verma A. First human experience with real-
by intracardiac echocardiography. Circulation 1995; 92:1839–1848. time integration of intracardiac echocardiography and 3D electro-
21. Verma A, Marrouche NF, Natale A. Pulmonary vein antrum anatomical imaging to guide right free wall accessory pathway
isolation: intracardiac echocardiography-guided technique. J Car- ablation. Europace 2008; 10:116–117.
diovasc Electrophysiol 2004; 15:1335–1340. 40. den Uijl DW, Tops LF, Tolosana JM, et al. Real-time integration of
22. Ren JF, Marchlinski FE. Utility of intracardiac echocardiography intracardiac echocardiography and multislice computed tomog-
in left heart ablation for tachyarrhythmias. Echocardiography raphy to guide radiofrequency catheter ablation for atrial fibrilla-
2007; 24:533–540. tion. Heart Rhythm 2008; 5:1403–1410.
[18/6/010/20:23:13] [515–525]
41. Nakagawa H, Lazzara R, Khastgir T, et al. Role of the tricuspid 53. Mitchell AR, Timperley J, Hudsmith L, et al. Intracardiac echo-
annulus and the eustachian valve/ridge on atrial flutter. Rele- cardiography to guide myocardial biopsy of a primary cardiac
vance to catheter ablation of the septal isthmus and a new tech- tumour. Eur J Echocardiogr 2007; 8:505–506.
nique for rapid identification of ablation success. Circulation 1996; 54. Segar DS, Bourdillon PD, Elsner G, et al. Intracardiac echocar-
94:407–424. diography-guided biopsy of intracardiac masses. J Am Soc Echo-
42. Kedia A, Hsu PY, Holmes J, et al. Use of intracardiac echocardiog- cardiogr 1995; 8:927–929.
raphy in guiding radiofrequency catheter ablation of atrial tachy- 55. Ho IC, Neuzil P, Mraz T, et al. Use of intracardiac echocardiog-
cardia in a patient after the senning operation. Pacing Clin raphy to guide implantation of a left atrial appendage occlusion
Electrophysiol 2003; 26:2178–2780. device (PLAATO). Heart Rhythm 2007; 4:567–571.
43. Callans DJ, Zado E, Sarter BH, et al. Efficacy of radiofrequency 56. Cao QL, Zabal C, Koenig P, et al. Initial clinical experience with
catheter ablation for ventricular tachycardia in healed myocardial intracardiac echocardiography in guiding transcatheter closure of
infarction. Am J Cardiol 1998; 82:429–432. perimembranous ventricular septal defects: feasibility and com-
44. Morady F, Harvey M, Kalbfleisch SJ, et al. Radiofrequency cath- parison with transesophageal echocardiography. Catheter Cardi-
eter ablation of ventricular tachycardia in patients with coronary ovasc Interv 2005; 66:258–267.
artery disease. Circulation 1993; 87:363–372. 57. Sorajja P, Valeti U, Nishimura RA, et al. Outcome of alcohol septal
45. Callans DJ, Ren JF, Michele J, et al. Electroanatomic left ventricular ablation for obstructive hypertrophic cardiomyopathy. Circulation
mapping in the porcine model of healed anterior myocardial 2008; 118:131–139.
infarction. Correlation with intracardiac echocardiography and 58. Alfonso F, Martin D, Fernandez-Vazquez F. Intracardiac echocar-
pathological analysis. Circulation 1999; 100:1744–1750. diography guidance for alcohol septal ablation in hypertrophic
46. Jongbloed MR, Bax JJ, van der Burg AE, et al. Radiofrequency obstructive cardiomyopathy. J Invasive Cardiol 2007; 19:E134–E136.
catheter ablation of ventricular tachycardia guided by intracardiac 59. Pedone C, Vijayakumar M, Ligthart JM, et al. Intracardiac echo-
echocardiography. Eur J Echocardiogr 2004; 5:34–40. cardiography guidance during percutaneous transluminal septal
47. Khaykin Y, Skanes A, Whaley B, et al. Real-time integration of 2D myocardial ablation in patients with obstructive hypertrophic
intracardiac echocardiography and 3D electroanatomical mapping cardiomyopathy. Int J Cardiovasc Intervent 2005; 7:134–137.
to guide ventricular tachycardia ablation. Heart Rhythm 2008; 60. Green NE, Hansgen AR, Carroll JD. Initial clinical experience with
5:1396–1402. intracardiac echocardiography in guiding balloon mitral valvulo-
48. Abraham WT, Fisher WG, Smith AL, et al. Cardiac resynchroni- plasty: technique, safety, utility, and limitations. Catheter Cardi-
zation in chronic heart failure. N Engl J Med 2002; 346:1845–1853. ovasc Interv 2004; 63:385–394.
49. Cazeau S, Leclercq C, Lavergne T, et al. Effects of multisite 61. Chessa M, Butera G, Carminati M. Intracardiac echocardiography
biventricular pacing in patients with heart failure and intraven- during percutaneous pulmonary valve replacement. Eur Heart J
tricular conduction delay. N Engl J Med 2001; 344:873–880. 2008; 29:2908.
50. Cleland JG, Daubert JC, Erdmann E, et al. The effect of cardiac 62. Naqvi TZ, Zarbatany D, Molloy MD, et al. Intracardiac echocar-
resynchronization on morbidity and mortality in heart failure. N diography for percutaneous mitral valve repair in a swine model.
Engl J Med 2005; 352:1539–1549. J Am Soc Echocardiogr 2006; 19:147–153.
51. Scholten MF, Szili-Torok T, Thornton AS, et al. Visualization of a 63. Horowitz BN, Vaseghi M, Mahajan A, et al. Percutaneous intra-
coronary sinus valve using intracardiac echocardiography. Eur J pericardial echocardiography during catheter ablation: a feasibil-
Echocardiogr 2004; 5:93–96. ity study. Heart Rhythm 2006; 3:1275–1282.
52. Shalaby AA. Utilization of intracardiac echocardiography to 64. Rodrigues AC, d’Avila A, Houghtaling C, et al. Intrapericardial
access the coronary sinus for left ventricular lead placement. echocardiography: a novel catheter-based approach to cardiac
Pacing Clin Electrophysiol 2005; 28:493–497. imaging. J Am Soc Echocardiogr 2004; 17:269–274.
[15/7/010/23:21:31] [526–540]
41
TEE to guide interventional cardiac procedures

in the catheterization laboratory
Matthias Greutmann, Melitta Mezody, and Eric Horlick
INTRODUCTION In practical terms, a major advantage of ICE is freedom from

Since its development for clinical use almost 30 years ago (1,2), additional personnel required to arrange a procedure (sonogra-
transesophageal echocardiography (TEE) has seen widespread pher and anesthetist). In many centers, the scarcity of anesthe-
use for diagnostic purposes and has been used to guide many tists remains a challenge in arranging complex procedures. In
new interventional cardiac procedures. More recently, intra- the pediatric structural laboratory where general anesthesia is
cardiac echocardiography (ICE) has begun to replace TEE for the rule, there is little role for ICE, and TEE remains the
some indications. However, the wide availability of TEE, the preferred imaging modality.
long-standing clinical experience with this technique, and its The current generation of ICE probes allows two-dimen-
well-documented safety profile (3,4) have preserved its impor- sional (2D) imaging including color and spectral Doppler
tant role as guidance for a variety of interventional procedures. techniques, while all modern TEE systems allow multiplane
This chapter discusses the role of TEE for guiding cardiac imaging with rapidly evolving technology for 3D imaging.
interventions with a special focus on the most commonly Although ICE allows high-quality visualization of most cardiac
performed procedures and on newer developments, such as structures, it remains mostly a tool for interventions on the
three-dimensional (3D) and real-time 3D echocardiography. atrial septum and the aortic root. To date, there is little expe-
rience with interventions in other locations within the heart,
and standardized protocols for its use are lacking. For example,
General Aspects no literature is available about how to quantify valvular regur-
Periprocedural echocardiographic guidance is most useful for
gitation or stenosis of left-sided valve lesions with ICE.
device closure of interatrial or interventricular communications
and for interventions on the left-sided (systemic) cardiac valves
3D Echocardiography
(5). In the peri-interventional setting, echocardiographic guid-
Initial experience with 3D and real-time 3D TEE for various
ance fulfills three fundamental roles in each single procedure:
procedures has been collected, and results are promising (8,9).
1. Confirms the indication and excludes contraindications 3D imaging improves the interventionalist’s perception of the
2. Guides the procedure to improve its safety and increase its location of the lesion to be treated, which is often lost with
success rate. omniplane 2D images. The interventionalist can gain an
3. Confirms early procedural efficacy and identifies immedi- improved appreciation of where wires and catheters are in
ate or imminent complications. the 3D space related to the target. By virtue of its improved
spatial resolution, 3D TEE has the potential to facilitate proce-
The role of TEE before, during, and after various non- dures in patients with complex cardiac anatomy or complex
coronary cardiac interventions is outlined in Table 41.1. There lesion geometry. 3D imaging usually adds little for the post-
is a large degree of variability in its use among institutions and procedural assessment. Full-volume 3D acquisitions require
individual interventionalists. extensive postprocessing (slice and dice technique) and are
therefore not useful to guide a procedure. Procedural guidance
TEE Vs. ICE by 3D TEE remains a domain of real-time image acquisition.
A detailed discussion of ICE for cardiac interventions is given in One drawback of real-time 3D is that 3D rendering is coarse
chapter 40. Compared to ICE, with equipment costs of many and slow. Future generations of technology and faster com-
thousands of dollars per procedure, the performance of TEE is puter speeds are likely to improve the experience significantly.
far less expensive. However, TEE requires general anesthesia New generations of ICE probes capable of 3D imaging are on
and those indirect costs also have to be taken into account. For the horizon as well. At present there is generally limited
interventionalists starting to perform structural interventions, it knowledge among echocardiographers with regard to optimal
is reassuring to have another expert imager in the catheter- utilization of these new techniques, which in itself serves as a
ization laboratory to help guide the procedure. That being said, limitation.
many experienced operators continue to rely on TEE. In many
laboratories, mainly depending on reimbursement, ICE has
begun to replace TEE for guiding septal procedures. ICE has Teamwork and Communication
been shown to lower procedure times and to be equivalent to When handling the ICE probe, the interventionalist can imme-
TEE in terms of procedural success and safety for closure of diately adjust views to his or her needs. In contrast, when
atrial septal defects (ASDs) (6,7). Its main advantage is TEE is used to guide a procedure, clear communication
improved patient comfort and freedom from general anesthesia. between echocardiographer and interventionalist is crucial.
[15/7/010/23:21:31] [526–540]
TEE TO GUIDE INTERVENTIONAL CARDIAC PROCEDURES IN THE CATHETERIZATION LABORATORY 527
Table 41.1 Value of Transesophageal Echocardiography for Commonly Performed Noncoronary Cardiac Interventions
During the Immediate
Procedure Preprocedure procedure postprocedure Comments
ASD þþþ þþþ þþþ 3D promising
Baffle leak closure þþþ þ þ TEE rarely used
PFO closure þþþ þa þa Not mandatory for the
procedure
VSD closure þþþ þþ þ 3D potentially helpful
Balloon aortic valve dilatation þþ þ þ Limited use, not mandatory
Aortic valve replacement þþ þþþ þþ Critical for stent placement
“Aortic root” interventions (i.e., ruptured sinus þþþ þþþ þþþ
Valsalva aneurysms)
Pulmonary valve interventions – – –
Transseptal puncture þ þþ þ Enhances safety
Balloon mitral valvuloplasty þþ þ þþ Not mandatory
Paravalvular leak closure þþþ þþþ þþ 3D promising
Interventional mitral valve repair þþþ þþþ þþþ
PDA closure – – –
Coarctation stenting – – –
–, not recommended for routine use, may be helpful in individual cases; þ, may be helpful in some instances; þþ, very helpful but not
mandatory; þþþ, mandatory/very helpful.
a
Use of TEE for PFO closure varies depending on institution.
Abbreviations: ASD, atrial septal defect; PDA, patent ductus arteriosus; PFO, patent foramen ovale; VSD, ventricular septal defect.
The echocardiographer should be familiar with the procedure, on image quality, being more often than not suboptimal in the
know the critical steps, and should understand what informa- adult population. 3D TEE technology has evolved rapidly since
tion is relevant to the interventionalist. The echocardiographer the late 1990s. It allows visualization of the changing geometry
should also be familiar with potential complications and should of ASDs throughout the cardiac cycle (Fig. 41.1) and allows
know what to expect and when to look for it. Using TEE for better definition of their borders as well as their relationship to
guiding interventions therefore requires optimal teamwork. As atrioventricular valves and venous inflows (20–24). Until
interventionalists and echocardiographers often speak in dif- recently, 3D reconstruction had to be made off-line and there-
ferent terms of reference, clear communication is important. In fore was not well suited to real-time procedure guidance. Only
our experience, it is of great advantage to have dedicated the recent evolution of real-time 3D TEE has allowed it to guide
interventionalists and echocardiographers who work together procedures in the catheterization laboratory (8). As it poten-
frequently. Teams who work well together use both verbal and tially allows a better understanding of the 3D geometry in cases
nonverbal communication to improve procedural efficacy. of distorted atrial septal geometry, it may have the potential to
facilitate successful device deployment. More experience and
Outlook comparative studies are required to further define its role.
In the following sections, the role of TEE for various interven-
tions is outlined, including detailed descriptions of selected
critical steps and a summary of key points in a checklist format TEE for Guidance of Secundum-Type ASD
for some of the most important and most common interventions. Closure—Practical Aspects
The technical details of closure of secundum-type ASDs are
outlined in detail elsewhere (see chap. 46). This section covers
the role of TEE during these procedures. Because of their ease
TEE FOR DEVICE CLOSURE OF INTERATRIAL
of use, high procedural success, and low complication rates,
COMMUNICATIONS
double-disc septal devices such as Amplatzer1 (AGA Medical
General Aspects
Corp., Plymouth, Minnesota, U.S.) and Occlutech1 (Helsing-
Device closure of interatrial communications is among the most
borg, Sweden) are currently the most frequently used septal
frequently performed noncoronary intervention in the cardiac
closure devices on the market. Other innovative device designs
catheterization laboratory. TEE has evolved as a reliable
are available as well and the general principles of device
method for guiding percutaneous closure of ASDs with con-
implantation for ASDs are similar for all designs. In this
temporary devices. Its usefulness for planning and guiding the
section, we will focus on the use of double-disc devices.
procedure has been demonstrated in numerous studies (10–17).
An overview of the role of TEE for these interventions is
Most procedures are guided by both fluoroscopy and TEE, but
given in Table 41.2, followed by a detailed description of critical
even guidance with TEE alone has shown to be feasible (18).
steps.
The Role of 3D Echocardiography Preprocedural

Although real-time transthoracic 3D echocardiography has Only defects confined to the oval fossa, so called secundum or
shown to be of increasing value for characterization of secun- “true” ASDs without anomalous drainage of pulmonary veins,
dum-type ASDs (19), TEE has remained the gold standard. The are amenable to device closure. The exclusion of a sinus
accuracy of transthoracic studies remains crucially dependent venosus defect, primum ASD, coronary sinus defect, and the
[15/7/010/23:21:31] [526–540]
Figure 41.1 (See color insert) Reconstruction of the “true” size and geometry of a secundum atrial septal defect derived from postprocessing
of a 3D transesophageal echocardiography acquisition (bottom right). Abbreviations: LA, left atrium; RA, right atrium.
Table 41.2 Transesophageal Echocardiography Checklist for Atrial Septal Defect Closure
Preprocedural
3 Rule out intracardiac thrombus in patients with atrial fibrillation
3 Assess left ventricular diastolic function and estimate left atrial pressure
3 Define size, anatomy, and location of the defect
3 Rule out sinus venosus, coronary sinus, and primum atrial septal defects
3 Rule out anomalous pulmonary venous drainage
3 Assess severity of tricuspid regurgitation and estimate right ventricular systolic pressure
3 Rule out other congenital or acquired cardiac lesions
3 Ensure adequate tissue rims toward AV valves, caval veins, and pulmonary veins (at least 5 mm with most Amplatzer1 devices)
3 Assess whether rims appear floppy or firm
3 Rule out or define multiple defects
3 Define presence of an atrial septal aneurysm
During the procedure
3 Confirm position of guidewire and delivery sheath
3 Confirm position of the sizing balloon and cessation of shunting on color Doppler when the sizing balloon is inflated
3 Measurement of sizing balloon waist (“stop flow” size)
3 Confirm proper alignment of left atrial disc to interatrial septum
3 Ensure permanent pacemaker wires are not entrapped
3 Confirm absence of entangling in Chiari network
3 Confirm absence of prolapse of left atrial disc at aortic margin
Postprocedural
3 Confirm proper position and alignment of the device, rule out device prolapse
3 Confirm absence of “rubbing” of the device against the aortic root
3 Confirm normal function of atrioventricular valves and absence of aortic regurgitation
3 Confirm unobstructed inflow of caval veins, right pulmonary veins, and coronary sinus
3 Confirm absence of pericardial effusion and signs of tamponade
[15/7/010/23:21:31] [526–540]
Figure 41.2 (See color insert) A 61-year-old patient with a large superior sinus venosus defect. Panel A: Mid-esophageal four-chamber
view, showing enlarged right-sided heart chambers. Panel B: Standard bicaval view of interatrial septum does not demonstrate superior sinus
venosus defect. Panel C: Only further withdrawing and slight clockwise rotation of the echo probe shows the large superior sinus venosus
defect (white arrow). Panel D: After injection of agitated saline into left antecubital vein immediate opacification of both atrial chambers from
the SVC (arrowheads). Abbreviations: RV, right ventricle; LV, left ventricle; LA, left atrium; RA, right atrium; SVC, superior vena cava.
demonstration of normal drainage of four pulmonary veins is Table 41.3 Most Useful Transesophageal Echocardiography
usually performed during an outpatient preinterventional Views for Visualization of an Atrial Septal Defect and Its Surround-
study. In case of incomplete assessment on those studies, a ing Structures
thorough examination at the time of the planned closure pro- TEE view (may vary in individual
cedure is mandatory (Fig. 41.2). Structure to visualize patients)
The Amplatzer septal occluder is currently available in Inferoposterior and 808–1108 “bicaval view”
sizes up to 40 mm and hence defects larger than 38 to 39 mm by superoposterior
balloon sizing are not amenable to device closure. Secundum (vena cava) rims
defects can extend in any direction, toward the orifices of the Posterior rim Mid-esophageal 0–308
superior or inferior vena cava and the coronary sinus, ante- Anterosuperior (aortic) rim Mid-esophageal 20–408
rosuperior toward the aortic root and toward the atrioventric- Distance to AV valves 0–408 Mid-esophageal 4-chamber
ular valves. Given the design of the double-disc devices, a view
tissue rim of at least 5 mm toward most of these structures is Distance to coronary sinus 08 View at esophago-gastric
mandatory for stable device positioning. A partially deficient junction
rim is the rule rather than the exception, and its characteriza-
tion is important for planning the procedure (25). The vast
majority of defects larger than 20 mm in size have an absent turning the probe slightly clockwise from a bicaval view.
aortic rim. Table 41.3 gives an overview of the most valuable Keeping the right upper pulmonary vein in view and slowly
TEE views to define tissue rims surrounding a defect. decreasing the plane angle toward 308 to 608 with slight clock-
The best angle to identify drainage of pulmonary veins is wise rotation of the probe, the right lower pulmonary vein can
highly variable and differs from patient to patient. We usually be identified. As an alternative approach, pulmonary veins can
start by identifying the left upper pulmonary vein at an angle of be identified from a 08 angle by gently advancing and with-
608 to 1108. It drains just above the left atrial appendage. By drawing the probe from a mid-esophageal view while turning
gently turning the probe counterclockwise and slightly increas- the probe either clockwise or counterclockwise. Fortunately, in
ing the angle, the left lower pulmonary vein is identified. The the presence of a significant left-to-right shunt across the
right upper pulmonary vein drains into the left atrium just interatrial septum, pulmonary vein flow is markedly increased
posterior to the superior vena cava and is easy to identify by and helps identifying these vessels. Finding only one
[15/7/010/23:21:31] [526–540]
pulmonary vein entering the left atrium on one side is not

abnormal as pulmonary veins are often confluent before enter-
ing the left atrium. However, a high suspicion for detection of
abnormal pulmonary venous drainage needs to be maintained,
especially if the dilatation of the right-sided heart chambers is
out of proportion to the size of the ASD.
During the Procedure (Procedural Aspects)

For device stability not only the size of tissue rims is important
but also whether they are characterized as firm or pliable. In the
case of an isolated absence of the anterosuperior or aortic rim
(Fig. 41.3), device closure is almost universally possible. A
stable device position can be achieved by “straddling” the
device around the aortic root. In these cases, the balloon should
be sized to the “stop flow” point—the balloon size at which
flow across the septum ceases—and not just to the point where
a waist appears on the balloon. Using a balloon to stretch the
septum should be avoided as it predisposes to selecting a larger
device that may increase the risk of erosion through the roof of
the atria and the aorta (26,27). In contrast to absent aortic rims,
a sufficient inferoposterior rim (Fig. 41.4) is crucial for a stable
position of the device. There is a trend to avoid balloon sizing
of ASDs before device implantation. Defect size is measured on
TEE, and device size is then arbitrarily selected at 4 to 5 mm
larger. Little evidence suggests that this practice is safer than
balloon sizing.
To rule out multiple defects it is crucial to interrogate the
interatrial septum with color Doppler while the sizing balloon
is inflated. If residual color flow is noted across the septum
while the sizing balloon is inflated, we would advocate entry of
the contralateral femoral vein to balloon size any substantial
secondary defect. The most important task of the echocardiog-
rapher during device closure of ASDs is assessment of the
deployed device and the detection of imminent complications
after implantation, but ideally before release of the device. We
Figure 41.3 Panel A–C: Mid-esophageal transesophageal echo-

cardiography views of an anterosuperior located secundum-type
atrial septal defect with absent aortic rim at different degree of
angulation. Large Eustachian valve (arrow), which must not be
mistaken for an inferoposterior tissue rim (arrowhead). Panel C
depicts a bicaval view and shows the true, in this case large, inferior Figure 41.4 (See color insert) Large atrial septal defect II (secun-
rim. Abbreviations: LA, left atrium; RA, right atrium; RAA, right atrial dum) with small posterior tissue rim (arrowhead) and good rim
appendage; AO, aorta. toward the AV valve level (arrow).
[15/7/010/23:21:31] [526–540]
spend adequate time verifying tissue margins in multiple Post Deployment

views, assessing residual shunting with color Doppler, and After deployment of the device, its position, any degree of
confirming the absence of impingement on surrounding struc- residual shunting, functional integrity of atrioventricular and
tures. It is prudent to be certain of device stability before its aortic valves, as well as unhindered systemic and pulmonary
release and to reposition or even change to another device size venous inflow must be documented. Thrombus formation on
if uncertain. Apart from device embolization, the most feared the left or right atrial disc has become rare with modern
complication is erosion of the roof of the atrium, which, one-third devices, proper use of full-dose heparinization, and restricting
of the time, involves the aortic root. A high level of awareness of the use of protamine. The frequency of thrombus occurrence
a developing pericardial effusion should be maintained. It is also with older-generation devices was up to 7% in early follow-up
important to exclude entrapment of a pacemaker wire during with TEE (30,31). After implantation it is very important to
device manipulation and final deployment. Some pacemaker carefully interrogate each margin for stability. Immediately
leads may be placed in a way that makes it impossible to avoid after the procedure, it is not uncommon to note leaking
wire entrapment, especially with large devices. This is not an through the device with color Doppler, which will cease
absolute contraindication for device placement, and the proce- with the expected thrombus formation between the disks. If
dure may be performed under special circumstances after careful color Doppler flow is noted at the superior vena cava or the
consideration of alternatives. aortic rim, the device should often be upsized by 2 mm for
Deployment of the left atrial disc in the left atrial append- devices >20 mm and by 1 to 2 mm for devices <20 mm. In
age or against the left atrial wall may result in damage to these large defects, it is common to find minimal residual leaks with
structures and should be avoided. Deploying the left atrial disc color Doppler at the inferior caval vein margin. When the
after rotation has occurred in the delivery system may lead to septum is thin, this residual color Doppler flow may be caused
“cobra head” malformation, which is easily detectable on TEE. by small fenestrations or Thebesian veins. If the inferior
To avoid this complication, it is crucial to guide retrieval of the septum, however, is seen within the “jaws” of the device,
delivery sheath by TEE into the middle of the left atrium before stability is further reassured.
deploying the left atrial disc of the device, especially if larger
devices are required. Although it is thought that only complete
withdrawal of the device and manual reshaping outside the
body could resolve cobra head malformation (28), in our expe-
Multiple Atrial Septal Defects or Fenestrated
rience, simply advancing the left atrial disc further into the left
Atrial Septum
Although multiple defects are no longer considered to be a
atrium usually restores normal device configuration.
contraindication for device closure, their recognition and
With large defects the left atrial disc can become perpen-
detailed delineation is of paramount importance for planning
dicular to the defect itself leading to prolapse at the aortic valve
and executing a successful intervention. Particularly in the
margin and preventing proper delivery of the device. This
presence of multiple defects, 3D imaging has proven to be
problem is less evident on fluoroscopy, but can be easily
helpful in understanding the spatial relationship of those
recognized on TEE. A second maneuver to resolve aortic disc
defects (Fig. 41.5) and hence planning of the intervention (21).
prolapse includes rotating the delivery system clockwise to the
Depending on the strategy chosen, it is important to ensure on
right side of the left atrium followed by usual deployment,
TEE that the guidewire passes through one of the central
which is often successful. In these cases, the use of a Hausdorf
defects or through the largest defect. However, it is often
sheath is advocated; however, we have never required one so
difficult to be certain of wire position in the presence of a
far in over 1200 cases. Opening the left atrial disc in the left
billowing aneurysmal septum. In those circumstances, we
upper pulmonary vein and rapidly exposing the right atrial
advocate crossing a defect arbitrarily with a wire and inflating
disc works well for devices >34 mm in size. After such a
a sizing balloon within this defect, before trying to cross
maneuver, confirming good alignment of the device toward
secondary defects with another catheter and guidewire. This
the interatrial septum and exclusion of prolapse of the device
technique requires a second venous puncture but avoids the
are important to minimize the risk of subsequent device
possibility of crossing a small defect and assuming it is dom-
embolization.
inant, leaving a larger defect to be dealt with by the disks of a
The StarFlex1 and BioStar1 devices have been used
device placed in a smaller hole.
infrequently for the closure of small ASDs. It is most important
to ensure that the anterior device arm of the distal umbrella does
not prolapse into the right atrium before releasing the proximal
umbrella, as this may lead to a significant residual shunt and Device Closure of Patent Foramen Ovale
increases the likelihood of device embolization. The best view to Echocardiographic guidance for device closure of patent fora-
ensure good position of the distal arms is a short-axis view men ovale is not mandatory (32,33). However, this issue is still
through the base of the heart (usually mid-esophageal 208–508). contentious and there is an ongoing debate about the safest way
Further tips and tricks for troubleshooting with this device are of closing a patent foramen ovale (PFO) (33,34). From a prac-
given in chapter 46. tical point of view, we, as others, use echocardiographic guid-
There are reports of entanglement of an ASD closure ance in addition to fluoroscopy only for those patients in whom
device in a Chiari network, a complication that should be easily complex geometry is anticipated. This includes PFOs with long
detected on TEE (29). This is a very rare complication and tunnels, large atrial septal aneurysms, or if new devices are
avoiding entanglement of pacemaker or ICD leads during used, especially those that cannot be easily retrieved once
insertion of the sheaths and when deploying the device is of deployed (35). If TEE or ICE is not used for PFO closure, the
far greater clinical importance. use of angiography post implantation is mandatory.
[15/7/010/23:21:31] [526–540]
Figure 41.5 (See color insert) Panels A to C depict transesophageal echocardiography images of a large secundum-type atrial septal defect
at an angle of 08, 1098, and 298 From 2D images the impression arises that there are two defects, but only the 3D reconstruction (view from
left atrium shows the true extent of the defect, being a common atrium with a narrow tissue bridge (arrow in panel D) across the atrium, rather
than two separate defects (arrowheads). Abbreviations: RA, right atrium; LA, left atrium.
Device Closure of Baffle Leaks After Atrial Switch 3D echocardiographic imaging may become more important in
and Fontan-Type Operations guiding these procedures, so far however, experience is limited.
Baffle leaks after atrial switch operations are common. They In contrast to baffle leaks after atrial switch operations,
behave as ASDs. There is either predominant left-to-right shunt residual fenestrations in the systemic venous limb of the Fontan
and associated volume load of the subpulmonic left ventricle or circulation are often deliberately created at the time of surgery
right-to-left shunt, typically in the presence of a concomitant to reduce immediate postoperative complications. This occurs
distal baffle obstruction. The latter leaves the patient cyanotic at the price of persistent right-to-left shunting. Many fenestra-
and at risk for paradoxical embolism. This is especially true in tions do not close spontaneously. Elective device closure of
the company of a transvenous pacemaker lead. As the anatomy these residual right-to-left shunts is therefore often performed
in these patients is often complex and distorted, a detailed when significant cyanosis persists. We rarely use TEE guidance
understanding of the anatomy is crucial to succeed (36). Full for these procedures but some cases have been reported where
visualization of atrial baffles after atrial switch procedures can it has been useful (40,41).
be difficult on TEE. To visualize the systemic venous baffles, one
practical approach is to find the mitral valve (at 08–408) and then
to follow the inferior and superior cava baffles by slowly with- TEE FOR DEVICE CLOSURE OF VENTRICULAR
drawing [superior vena cava (SVC)] and advancing [inferior SEPTAL DEFECTS
vena cava (IVC)] the probe, while simultaneously rotating gen- General Aspects
tly clockwise. Sometimes adequate assessment of the systemic Device closure of congenital muscular ventricular septal defects
venous baffles can also be achieved from a modified bicaval (VSD) or residual VSDs after open heart surgery is a well-
view. The pulmonary venous baffle is posterior to the systemic established procedure with a low complication rate and a high
baffle. Pulmonary venous baffle obstructions are much less rate of procedural success. In contrast, device closure of peri-
frequent. Visualization of pulmonary venous drainage is com- membranous VSDs is associated with a risk of complete heart
parable to visualization of pulmonary veins in the normal heart. block of up to 5% in a mostly pediatric population. Device
Bubble contrast echocardiography is the most sensitive way for closure is therefore more contentious in patients at low risk for
detection of even small right-to-left shunts. The role of TEE to elective surgical repair and especially in children. Finally,
guide the closure of baffle leaks is limited but in some instances postmyocardial infarction VSDs are associated with a high
may be helpful for defining the optimal treatment strategy (37– mortality rate both when left untreated and when treated
39). The ability to delineate spatial relationships with real-time successfully with device closure in the acute setting. Device
[15/7/010/23:21:31] [526–540]
Table 41.4 Transesophageal Echocardiography Checklist for (usually 708–1008), and long-axis views (1008–1358) as well as
Ventricular Septal Defect Device Closure transgastric short-axis views and deep transgastric four-cham-
ber views (both at 08). It is important to visualize a defect in
Preprocedural
several planes to appreciate its size and geometry. In addition,
3 Define size, anatomy, exact location, and number of defects
(multiple views) for perimembranous VSDs, a short-axis view through the base
3 Rule out aortic valve cusp prolapse and malalignment VSD of the heart (208–608) is helpful. As apically located VSDs may
3 Rule out additional congenital cardiac defects be hard to visualize on TEE, complementary transthoracic
3 Define relation and distance to AV valves and aortic valve imaging might be helpful in these cases; we generally do not
During the procedure use TEE guidance in patients with apical defects.
3 Confirm position of guidewire and delivery sheath
3 Confirm position and device stability before release (multiple 3D Echocardiography
views) As for other indications, 3D echocardiography is particularly
3 Assess residual shunting before device release useful for delineation of spatial relationship between multiple
Postprocedural defects and hence might help planning of optimal interven-
3 Confirm proper position and alignment of the device tional or surgical strategies (44).
3 Document any degree of residual shuntinga
3 Exclude significant aortic regurgitation/distortion of aortic rootb Postinfarction VSDs
a
Color Doppler tends to underestimate the degree of a residual VSDs after myocardial infarction differ in many ways from
shunt and one needs to be cautious not to overestimate the success congenital VSDs. Interventionalists can be involved in the care
of a given procedure. Even small residual shunts may predispose to of these patients when they present either as acute postinfarc-
hemolysis. tion VSDs or when patients have residual defects after surgical
b
Most important for device closure of perimembranous VSDs. closure of a postinfarction VSD. The outlook for the former is
Abbreviation: VSD, ventricular septal defect.
bleak when left untreated, and every intervention, either sur-
gical or interventional, remains at very high risk. Those treated
after “a trial of life” in the subacute phase often have a much
closure of postoperative or subacute postmyocardial infarction improved outlook; however, if patients deteriorate during their
VSDs is usually associated with acceptable outcomes. trial, late intervention is usually at even greater risk. Rather
In many series, VSD device closure was performed under than being well-defined defects, postinfarction VSDs are often
fluoroscopic and TEE guidance. Fluoroscopic guidance alone or jagged edged, irregular, and serpiginous ruptures of the inter-
guidance by fluoroscopy and transthoracic echocardiography ventricular septum with poorly defined borders. They often
was used in others, especially for single muscular VSDs (42,43). have multiple exit points toward the right ventricle. To allow
In patients with complex anatomy, for example, residual post- proper planning of interventional closure, it is therefore crucial
operative shunts after complex repair of congenital heart dis- to define defect size, location, borders, and inflow and out-
ease, TEE usually shares the same obstacles and difficulties as flows.
angiography, namely atypical imaging planes. Shadowing arti- As the largest dedicated double-disc device size at pres-
facts created by calcified VSD patches from previous surgical ent is 24 mm, proper sizing of the defect is critical. Apart from
repair or prosthetic valves can hamper image quality signifi- defining defect location and morphology, echocardiography
cantly. The utility of TEE can only be assessed on a case-by-case defines left ventricular function, integrity of the valvular appa-
basis. It may be helpful in patients with multiple defects or in ratus, and defect location (Fig. 41.6). It is important to detect
those with defects in proximity to other cardiac structures such pericardial effusion, as this may herald free wall or contained
as atrioventricular and aortic valves. A peri-interventional rupture. The location of the defect with respect to the rims is
diagnostic study often clarifies this. Sizing of VSDs is per- quite critical, as one tries to avoid putting stress on a structur-
formed by angiography and TEE at end-diastole or with a ally weakened septum to free wall junction, which might result
compliant sizing balloon (Table 41.4). in free wall rupture. As there is often more than one defect, it is
also crucial to assure that at time of the intervention, the
Practical Aspects guidewire is crossing the largest defect (45–47).
Detailed preprocedural echocardiography is important not only
for defining defect size, location, and relationship to the sur-
rounding structures but also for planning the technical details TEE FOR INTERATRIAL TRANSSEPTAL
of a procedure. If the defect is located anterior or toward the PUNCTURE
outlet septum, a transfemoral approach is preferred, while in Practical Aspects
the past, for most other defect localizations, a transjugular Access to the left atrium from a systemic venous path is a
access was chosen. In the adult heart it is, however, easy to prerequisite for many cardiac interventions. This includes
form a loop in the right atrium, especially with the availability percutaneous mitral valve repair, mitral balloon valvuloplasty,
of braided sheaths, which mimic a jugular orientation of guide- device closure of left atrial appendage, device closure of para-
wire and delivery sheath. This allows the performance of most valvular leaks, and many electrophysiological procedures (48).
VSD closures from a femoral approach. If a muscular Amplat- Although echocardiographic guidance is not mandatory, it
zer septal occluder is used, the distance to atrioventricular and provides additional safety and helps to localize the optimal
aortic valves must be at least 4 mm. The asymmetric Amplatzer site of puncture for a specific intervention. Different procedures
perimembranous VSD occluder allows implantation if the dis- require puncture at different sites in the interatrial septum, for
tance to the aortic valve is as little as 1 to 2 mm. example, a high posterior puncture when a mitral valve clip
The best views on TEE to define a muscular VSD are the repair is planned (49,50). Younger interventionalists are
mid-esophageal four-chamber (usually 08–208), two-chamber increasingly dependent on TEE guidance as the need for
[15/7/010/23:21:31] [526–540]
3D Echocardiography
Real-time 3D imaging has recently been shown to be helpful for
some interventions. Because of its unique ability to delineate
spatial relationship, it might prove useful for guiding trans-
septal puncture in certain patients with high-risk anatomy. This
includes patients with congenital heart disease, pulmonary
artery hypertension, left to right bowing of the septum in the
presence of mitral stenosis, or in cases of a thickened lip-
omatous septum (54).
TEE FOR LEFT-SIDED HEART VALVE

PROCEDURES
Balloon Aortic Valvuloplasty
The role of TEE in guiding aortic valve balloon dilatation is
limited. When used, its purpose is accurate sizing of the aortic
annulus to choose the appropriate balloon size and to detect
immediate complications after each dilatation. Increasing aortic
regurgitation is prohibitive for further dilatations. Should the
Figure 41.6 (See color insert) An 84-year-old patient in cardio- patient become hypotensive following dilatation, TEE provides
genic shock due to subacute postmyocardial infarction ventricular an excellent tool to immediately detect or rule out complica-
septal defect. The arrow points to the left ventricular disc of an tions such as massive aortic regurgitation, aortic dissection,
Amplatzer1 ventricular septal defect occluder (white arrow), entan- tamponade, or new wall motion abnormalities suggestive of
gling the subvalvular mitral valve apparatus, leading to severe mitral coronary emboli.
regurgitation (arrowhead). Abbreviations: LA, left atrium; LV, left
ventricle.
Catheter-Based Aortic Valve Replacement
If catheter-based aortic valve replacement—either by the trans-
femoral or the transapical approach—is performed under gen-
transseptal puncture has become less frequent in the last two
eral anesthesia, TEE is usually used together with angiography
decades due to a decrease in rheumatic mitral valve disease
to guide the interventions. Alternatively, the transfemoral pro-
and the relative rarity of transseptal puncture for diagnostic
cedure—particularly with the Corevalve Revalving System
purposes (e.g., assessment of mitral or aortic valve disease).
(Medtronic CV, Luxembourg, E.U.)—can be performed in
Proper position of the transseptal needle is confirmed by
local anesthesia under pure angiographic guidance. Independ-
“tenting” of the interatrial septum with the needle withdrawn
ently of that, accurate sizing of the aortic annulus is critical, as
into the tip of the Mullins (Medtronic, Minneapolis, Minnesota,
the available prosthetic valve sizes are limited and some
U.S.) sheath. The puncture of the septum can be confirmed on
patients are disqualified for the procedure on the basis of
TEE by injection of agitated saline contrast through the trans-
their annulus size. TEE has been shown to be more accurate
septal needle, though more often radiographic contrast is used
than transthoracic echocardiography in sizing the annulus,
for this purpose (51,52). Transseptal puncture across patches,
especially in patients with heavily calcified valves (55,56).
baffles, or conduits in patients with repaired congenital heart
In addition to the sizing of the annulus, the main role for
disease is particularly challenging, and optimal delineation of
TEE is the optimal centering of the valved stent relative to the
an individual’s cardiac anatomy is important to enhance the
native aortic annulus (Fig. 41.7). With the transfemoral
safety of this step of the procedure (53) (Table 41.5).
approach using the balloon-expandable Edwards SAPIEN
valve, the aim is for a ventricular-to-aortic ratio of the stent of
Table 41.5 Transesophageal Echocardiography Checklist for 60:40—that is, the larger part of the stent before implantation
Interatrial Transseptal Puncture sits in the left ventricular outflow tract (LVOT)—as a slight
“travel” of the stent during deployment is expected. With the
Preprocedural
3 Define optimal site of puncture (usually in the region of the oval transapical approach less aortic travel is observed and therefore
fossa) the aim for centering is 50:50. Communication between the
3 Define anatomical distortion due to congenital heart disease, interventionalist and the echocardiographer during the proce-
mitral valve disease, or pulmonary hypertension dure is the key to success. We have used the term “ventricular”
3 Define localization of the aortic root and “aortic” to describe the position of the valve instead of “in
3 Rule out clot in left atrium or left atrial appendage and out” to avoid any ambiguity during this critical step (57).
3 Define presence and extent of lipomatous interatrial septum When using the transapical approach, TEE is even more impor-
During the procedure tant, as fluoroscopy in many operating rooms is provided by a
3 Confirm position of transseptal needle on the interatrial septum portable C-arm with less optimal image quality. In patients in
(tenting) whom a transapical approach is chosen, entanglement of the
3 Confirm intra-atrial location of needle tip after puncture (saline delivery system in the mitral valve apparatus can be life-
contrast) threatening and must be immediately detected by TEE. After
Postprocedural deployment of the stent, TEE should confirm stable stent posi-
3 Rule out pericardial effusion or tamponade tion and define the amount of paravalvular leakage. In cases of
3 Define size of residual atrial septal defect severe paravalvular regurgitation, postdilatation with a slightly
[15/7/010/23:21:31] [526–540]
Figure 41.7 (See color insert) Catheter-based aortic valve implantation. Panels A to D depict long-axis views of the aortic root. Panel A:
Placement of the valved stent (arrowheads mark edges of stent balloon; double-headed arrows mark the actual length of the valved stent).
Panel B: Valve implantation, balloon inflated. Panel C: Immediate post-valve deployment with thin prosthetic valve cusps visible. Panel D:
Short-axis view through base of the heart, depicting mild valvular and three small paravalvular jets (white stars) of aortic regurgitation.
Abbreviations: LA, left atrium; LVOT, left ventricular outflow tract; AO, aorta.
Table 41.6 Transesophageal Echocardiography Checklist for aneurysms) are often amenable to device closure. In an indi-
Catheter-Based Aortic Valve Replacement vidual patient it always has to be decided whether device
closure is expected to be equivalent to surgery, taking into
Preprocedural
account the patient’s estimated perioperative risk. Figure 41.8
3 Rule out mobile clot or active endocarditis
3 Rule out plaques grade IV or mobile clots in the thoracic aorta depicts the successful closure of a postoperative aortic to left
3 Accurate measurement of aortic annulus (between hinge points atrial fistula after bioprosthetic aortic valve replacement, and
of the valve) Figure 41.9 describes an attempted device closure of a ruptured
During the procedure sinus Valsalva aneurysm. TEE is very helpful in exact delinea-
3 Avoid entanglement of delivery system in mitral valve apparatus tion of a patient’s lesion anatomy, and the most useful views
in case of transapical approach are generally mid-esophageal short-axis view through the aor-
3 Confirm optimal position of valved stent before deployment tic root and long-axis views at 1108 to 1408. Sometimes, how-
Postprocedural ever, off-axis views may be required, especially for facilitating
3 Confirm stable position of the valved stent guidewire and sheath placement across a defect. Special note
3 Confirm normal prosthetic valve function should be made of multiple perforations of these aneurysms,
3 Assess severity and localization of paravalvular leaks which may compromise the result.
3 Assess immediate complications (new wall motion abnormalities
may help localizing the site of coronary obstruction)
Balloon Valvuloplasty of the Mitral Valve
The most important role of TEE in balloon valvuloplasty of the
mitral valve is the assessment of valve characteristics predict-
larger balloon volume might be efficacious. If the prosthesis is
ing the success of the procedure (thickening, calcification,
placed too low, a second valve may need to be deployed to
mobility of valve, and subvalvular apparatus) and is discussed
correct for the resulting severe aortic regurgitation (Table 41.6).
elsewhere in detail (see chap. 43). The absence of left atrial
thrombus is a prerequisite for a safe procedure and must be
Sinus Valsalva Aneurysm and Other confirmed by TEE even in patients with therapeutic anticoagu-
Aortic-to-Atrial Communications lation or in those in sinus rhythm (58).
Congenital or acquired communications between the aorta During the procedure itself, TEE guidance is not manda-
and various cardiac structures (i.e., ruptured sinus Valsalva tory but may ease transseptal puncture. It allows assessment of
[15/7/010/23:21:31] [526–540]
the severity of mitral regurgitation and other complications

immediately after the procedure (59). In addition, compared
with fluoroscopy alone, TEE guidance might improve place-
ment of the dilatation balloon and decrease procedure and
fluoroscopy time (60).
Device Closure of Paravalvular Leaks After

Prosthetic Heart Valve Surgery
Paravalvular leaks after surgical prosthetic heart valve replace-
ment can cause significant morbidity due to residual regurgi-
tation or hemolysis. To avoid the complications of a
reoperation, device closure has become an attractive alternative
to conventional surgery in poor surgical candidates. For these
technically demanding procedures, TEE plays an important
role for planning and executing these interventions. A detailed
description of the defect localization, size, and relationship to
adjacent structures is of paramount importance (61–63). Again
of great importance is a clear communication between the
echocardiographer and the interventionalist as localization on
echocardiography may not be easily translated into a biplane
radiographic view. Because of its unique ability to delineate
spatial relationships within the heart and its ability to visualize
the true geometry of a defect, real-time 3D TEE will likely
become the modality of choice for guidance of these procedures
in the future (Fig. 41.10). The application of real-time 3D TEE to
this intervention has provided tremendous improvement in
periprocedural success and sustainability. During the proce-
dure, TEE provides invaluable assistance to help target the
defect with a guidewire and delivery sheaths in 3D space
(Table 41.7).
RIGHT-SIDED VALVE LESIONS

The role of TEE for guiding right-sided heart valve procedures
is limited. Balloon dilatation of native pulmonary valve sten-
oses and implantation of percutaneous pulmonary valves is
usually performed under fluoroscopy alone, and the additional
yield of echocardiography is small. For optimal positioning of
the stent valve in the case of percutaneous treatment of a
stenosed or regurgitant conduit, calcifications within the wall
of the conduit are the most important landmarks for placement
of the device. The almost inevitably present heavy calcification
of these conduits hampers image quality on TEE significantly
because of shadowing artifacts. Interventional procedures on
the tricuspid valve are rarely performed and the role of TEE is
yet to be defined.
MISCELLANEOUS AND “FREESTYLE”

INTERVENTIONS
As many evolving procedures in the cardiac catheterization lab-
Figure 41.8 (See color insert) Device closure of fistula from aorta oratory, such as percutaneous mitral valve repair (see chap. 43),
to left atrium after bioprosthetic aortic valve replacement. Panel A: are critically dependent on optimal imaging, which is provided
Mid-esophageal short-axis view through aortic root. The arrowhead by TEE (64), its role will be increasing. The emerging technol-
points to the aortic valve prosthesis, the large arrow to the shunt. ogy of 3D TEE will even enhance its importance in the future.
Panel B: Angiographic view of sizing balloon (white star) measuring To cover all these interventions in detail is beyond the scope of
defect size. Panel C: Small residual left shunt after deployment of an this chapter and we therefore refer to the corresponding
Amplatzer duct occluder (arrow). Abbreviations: LA, left atrium; RA, chapters dealing with individual interventions. Whether TEE
right atrium; RV, right ventricle. or echocardiography is at all helpful for a particular interven-
tion is determined by several factors. First, if diagnostic TEE
[15/7/010/23:21:31] [526–540]
Figure 41.9 (See color insert) Attempted device closure of ruptured sinus Valsalva aneurysm. Panel A and B: Mid-esophageal short-axis
view through aortic root without and with color Doppler. The arrow points to the perforation of the sinus Valsalva aneurysm toward the right
atrium. Panel C: Angiographic view of the sinus Valsalva aneurysm. Panel D: Balloon sizing of the defect (white star). Panel E: Amplatzer duct
occluder sealing the perforation of the sinus Valsalva aneurysm (not released). The device seals only one of three communications between
the aorta and the right ventricle, leaving a large residual shunt. The device was therefore retrieved and the patient was referred for surgical
repair. Abbreviations: LA, left atrium; RA, right atrium; RV, right ventricle; AO, aorta.
Figure 41.10 (See color insert) Device closure of paravalvular mitral leak (bioprosthesis). Panels A and B: 748 Mid-esophageal
transesophageal echocardiography views demonstrating paravalvular mitral regurgitation (arrow). Panel C: Live 3D image of the mitral
valve prosthesis, seen from the left atrium demonstrating a large inferomedial paravalvular leak (arrow). The arrowhead points to the delivery
sheath crossing the leak. Panel D: Live 3D image after device closure of paravalvular leak with two Amplatzer1 duct occluder (white stars).
Abbreviations: LA, left atrium; LVOT, left ventricular outflow tract; AO, aorta.
[15/7/010/23:21:31] [526–540]
Table 41.7 Transesophageal Echocardiography Checklist for 3. Daniel WG, Erbel R, Kasper W, et al. Safety of transesophageal
Device Closure of Paravalvular Leaks echocardiography. A multicenter survey of 10,419 examinations.
Circulation 1991; 83:817–821.
Preprocedural 4. Min JK, Spencer KT, Furlong KT, et al. Clinical features of
3 Define localization, size, and geometry of the defect complications from transesophageal echocardiography: a single-
3 Define relationship to adjacent structures (heart valves, center case series of 10,000 consecutive examinations. J Am Soc
coronary arteries) Echocardiogr 2005; 18:925–929.
3 Rule out intracardiac clot and malfunction of the prosthetic valve 5. Cheung YF, Leung MP, Lee J, et al. An evolving role of trans-
3 Rule out “rocking” of the valve and confirm stability of the esophageal echocardiography for the monitoring of interventional
prosthesis catheterization in children. Clin Cardiol 1999; 22:804–810.
During the procedure 6. Boccalandro F, Baptista E, Muench A, et al. Comparison of intra-
3 Guide transseptal puncture cardiac echocardiography versus transesophageal echocardiogra-
3 Ease proper guidewire position (to AVOID entangling in phy guidance for percutaneous transcatheter closure of atrial
mechanical valves) septal defect. Am J Cardiol 2004; 93:437–440.
3 Rule out prosthesis malfunction caused by the closure device— 7. Mullen MJ, Dias BF, Walker F, et al. Intracardiac echocardiogra-
BEFORE release phy guided device closure of atrial septal defects. J Am Coll
Cardiol 2003; 41:285–292.
Postprocedural
8. Balzer J, Kuhl H, Franke A. Real-time three-dimensional trans-
3 Confirm normal prosthetic heart valve function
oesophageal echocardiography for guidance of atrial septal defect
3 Detect immediate complications
closures. Eur Heart J 2008; 29:2226.
3 Define the amount of residual paravalvular leakage and assess
9. Balzer J, Kuhl H, Rassaf T, et al. Real-time transesophageal
whether a second device is necessary
three-dimensional echocardiography for guidance of percutane-
3 Define the size of the residual atrial septal defect
ous cardiac interventions: first experience. Clin Res Cardiol 2008;
97:565–574.
10. Carminati M, Giusti S, Hausdorf G, et al. A European multicentric
experience using the CardioSEal and Starflex double umbrella
devices to close interatrial communications holes within the oval
was unable to visualize the structure of interest in high quality, fossa. Cardiol Young 2000; 10:519–526.
any yield in adding this modality is to be expected. This is the 11. Chan KC, Godman MJ. Morphological variations of fossa ovalis
case for any structure remote from the esophagus, as, for atrial septal defects (secundum): feasibility for transcutaneous
closure with the clam-shell device. Br Heart J 1993; 69:52–55.
example, a patent ductus arteriosus, aortopulmonary collater-
12. Chan KC, Godman MJ, Walsh K, et al. Transcatheter closure of
als, or peripheral pulmonary artery stenoses. Some clinical atrial septal defect and interatrial communications with a new self
scenarios preclude the use of TEE, including patients with expanding nitinol double disc device (Amplatzer septal occluder):
esophageal disease (e.g., strictures, varices) and those in multicentre UK experience. Heart 1999; 82:300–306.
whom it is safer to perform a procedure without general anes- 13. Cooke JC, Gelman JS, Harper RW. Echocardiologists’ role in the
thesia. The latter may be the case in hemodynamically unstable deployment of the Amplatzer atrial septal occluder device in
patients or patients with severe respiratory compromise. In adults. J Am Soc Echocardiogr 2001; 14:588–594.
those cases we avoid the use of TEE as we feel discouraged 14. Ferreira SM, Ho SY, Anderson RH. Morphological study of defects
about the safety of TEE with topical anesthesia and sedation of the atrial septum within the oval fossa: implications for
only. However, TEE without general anesthesia is a widely transcatheter closure of left-to-right shunt. Br Heart J 1992;
67:316–320.
practiced technique in many European centers. Intracardiac
15. Hellenbrand WE, Fahey JT, McGowan FX, et al. Transesophageal
echo might be more helpful in cases of this nature. echocardiographic guidance of transcatheter closure of atrial
Some lesions, such as pulmonary artery stenoses and septal defect. Am J Cardiol 1990; 66:207–213.
coarctation of the aorta, are easily and fully visualized by 16. Masura J, Gavora P, Formanek A, et al. Transcatheter closure of
angiography and therefore echocardiography is not routinely secundum atrial septal defects using the new self-centering
used for guiding these procedures. Its role remains limited to amplatzer septal occluder: initial human experience. Cathet Car-
immediate assessment of acute catastrophic complications, diovasc Diagn 1997; 42:388–393.
such as aortic dissection and tamponade. 17. Mazic U, Gavora P, Masura J. The role of transesophageal echo-
cardiography in transcatheter closure of secundum atrial septal
defects by the Amplatzer septal occluder. Am Heart J 2001; 142:
SUMMARY AND CONCLUSIONS 482–488.
The introduction of TEE into clinical practice has not only 18. Ewert P, Berger F, Daehnert I, et al. Transcatheter closure of atrial
septal defects without fluoroscopy: feasibility of a new method.
supported the development of many innovative cardiac inter-
Circulation 2000; 101:847–849.
ventions but has become an essential tool for guiding many of
19. van den Bosch AE, Ten Harkel DJ, McGhie JS, et al. Characteriza-
them. For some interventions ICE will likely replace TEE in the tion of atrial septal defect assessed by real-time 3-dimensional
future but the advent of real-time 3D TEE may lead to an echocardiography. J Am Soc Echocardiogr 2006; 19:815–821.
extension of its role in others. 20. Acar P, Saliba Z, Bonhoeffer P, et al. Influence of atrial septal
defect anatomy in patient selection and assessment of closure with
the Cardioseal device; a three-dimensional transoesophageal
REFERENCES echocardiographic reconstruction. Eur Heart J 2000; 21:573–581.
1. Hisanaga K, Hisanaga A, Nagata K, et al. Transesophageal cross- 21. Cao Q, Radtke W, Berger F, et al. Transcatheter closure of multiple
sectional echocardiography. Am Heart J 1980; 100:605–609. atrial septal defects. Initial results and value of two- and three-
2. Schluter M, Langenstein BA, Polster J, et al. Transoesophageal dimensional transoesophageal echocardiography. Eur Heart J
cross-sectional echocardiography with a phased array transducer 2000; 21:941–947.
system. Technique and initial clinical results. Br Heart J 1982; 22. Franke A, Kuhl HP, Rulands D, et al. Quantitative analysis of the
48:67–72. morphology of secundum-type atrial septal defects and their
[15/7/010/23:21:31] [526–540]
dynamic change using transesophageal three-dimensional echo- 42. Butera G, Chessa M, Carminati M. Percutaneous closure of
cardiography. Circulation 1997; 96:323–327. ventricular septal defects. State of the art. J Cardiovasc Med
23. Maeno YV, Benson LN, Boutin C. Impact of dynamic 3D trans- (Hagerstown) 2007; 8:39–45.
oesophageal echocardiography in the assessment of atrial septal 43. Hijazi ZM, Hakim F, Al-Fadley F, et al. Transcatheter closure of
defects and occlusion by the double-umbrella device (Cardio- single muscular ventricular septal defects using the amplatzer
SEAL). Cardiol Young 1998; 8:368–378. muscular VSD occluder: initial results and technical considera-
24. Magni G, Hijazi ZM, Pandian NG, et al. Two- and three-dimen- tions. Catheter Cardiovasc Interv 2000; 49:167–172.
sional transesophageal echocardiography in patient selection and 44. Acar P, Abdel-Massih T, Douste-Blazy MY, et al. Assessment of
assessment of atrial septal defect closure by the new DAS-Angel muscular ventricular septal defect closure by transcatheter or
Wings device: initial clinical experience. Circulation 1997; 96: surgical approach: a three-dimensional echocardiographic study.
1722–1728. Eur J Echocardiogr 2002; 3:185–191.
25. Podnar T, Martanovic P, Gavora P, et al. Morphological variations 45. Ahmed TA, Dare EV, Hincke M. Fibrin: a versatile scaffold for
of secundum-type atrial septal defects: feasibility for percutaneous tissue engineering applications. Tissue Eng Part B Rev 2008;
closure using Amplatzer septal occluders. Catheter Cardiovasc 14:199–215.
Interv 2001; 53:386–391. 46. Holzer R, Balzer D, Amin Z, et al. Transcatheter closure of post-
26. Amin Z, Hijazi ZM, Bass JL, et al. Erosion of Amplatzer infarction ventricular septal defects using the new Amplatzer
septal occluder device after closure of secundum atrial septal muscular VSD occluder: results of a U.S. Registry. Catheter
defects: review of registry of complications and recommendations Cardiovasc Interv 2004; 61:196–201.
to minimize future risk. Catheter Cardiovasc Interv 2004; 63: 47. Lock JE, Block PC, McKay RG, et al. Transcatheter closure of
496–502. ventricular septal defects. Circulation 1988; 78:361–368.
27. Divekar A, Gaamangwe T, Shaikh N, et al. Cardiac perforation 48. Roelke M, Smith AJ, Palacios IF. The technique and safety of
after device closure of atrial septal defects with the Amplatzer transseptal left heart catheterization: the Massachusetts General
septal occluder. J Am Coll Cardiol 2005; 45:1213–1218. Hospital experience with 1,279 procedures. Cathet Cardiovasc
28. Cooke JC, Gelman JS, Harper RW. Cobrahead malformation of the Diagn 1994; 32:332–339.
Amplatzer septal occluder device: an avoidable compilation of 49. Hahn K, Gal R, Sarnoski J, et al. Transesophageal echocardio-
percutaneous ASD closure. Catheter Cardiovasc Interv 2001; graphically guided atrial transseptal catheterization in patients
52:83–85; discussion 6–7. with normal-sized atria: incidence of complications. Clin Cardiol
29. Cooke JC, Gelman JS, Harper RW. Chiari network entanglement 1995; 18:217–220.
and herniation into the left atrium by an atrial septal defect 50. Tucker KJ, Curtis AB, Murphy J, et al. Transesophageal echocar-
occluder device. J Am Soc Echocardiogr 1999; 12:601–603. diographic guidance of transseptal left heart catheterization dur-
30. Krumsdorf U, Ostermayer S, Billinger K, et al. Incidence and ing radiofrequency ablation of left-sided accessory pathways in
clinical course of thrombus formation on atrial septal defect and humans. Pacing Clin Electrophysiol 1996; 19:272–281.
patient foramen ovale closure devices in 1,000 consecutive 51. Clark J, Bockoven JR, Lane J, et al. Use of three-dimensional
patients. J Am Coll Cardiol 2004; 43:302–309. catheter guidance and trans-esophageal echocardiography to
31. Moore JW, Levi DS. Transcatheter closure of atrial shunts. Focus eliminate fluoroscopy in catheter ablation of left-sided accessory
on a lingering issue. J Am Coll Cardiol 2004; 43:310–312. pathways. Pacing Clin Electrophysiol 2008; 31:283–289.
32. Varma C, Benson LN, Warr MR, et al. Clinical outcomes of patent 52. Kantoch MJ, Frost GF, Robertson MA. Use of transesophageal
foramen ovale closure for paradoxical emboli without echocardio- echocardiography in radiofrequency catheter ablation in children
graphic guidance. Catheter Cardiovasc Interv 2004; 62:519–525. and adolescents. Can J Cardiol 1998; 14:519–523.
33. Wahl A, Kunz M, Moschovitis A, et al. Long-term results after 53. El-Said HG, Ing FF, Grifka RG, et al. 18-year experience with
fluoroscopy-guided closure of patent foramen ovale for secondary transseptal procedures through baffles, conduits, and other intra-
prevention of paradoxical embolism. Heart 2008; 94:336–341. atrial patches. Catheter Cardiovasc Interv 2000; 50:434–439; dis-
34. Shishehbor MH, Christofferson RD, Tuzcu EM, et al. Long-term cussion 40.
results after PFO closure. Heart 2008; 94:100; author reply 1. 54. Chierchia GB, Capulzini L, de Asmundis C, et al. First experience
35. Hildick-Smith D, Behan M, Haworth P, et al. Patent foramen ovale with real-time three-dimensional transoesophageal echocardiog-
closure without echocardiographic control: use of “standby” intra- raphy-guided transseptal in patients undergoing atrial fibrillation
cardiac ultrasound. J Am Coll Cardiol Intv 2008; 1:387–391. ablation. Europace 2008; 10:1325–1328.
36. Balzer DT, Johnson M, Sharkey AM, et al. Transcatheter occlusion 55. Abraham TP, Kon ND, Nomeir AM, et al. Accuracy of trans-
of baffle leaks following atrial switch procedures for transposition esophageal echocardiography in preoperative determination of
of the great vessels (d-TGV). Catheter Cardiovasc Interv 2004; aortic annulus size during valve replacement. J Am Soc Echocar-
61:259–263. diogr 1997; 10:149–154.
37. Daehnert I, Hennig B, Wiener M, et al. Interventions in leaks and 56. Guarracino F, Zussa C, Polesel E, et al. Influence of transesopha-
obstructions of the interatrial baffle late after Mustard and Sen- geal echocardiography on intraoperative decision making for
ning correction for transposition of the great arteries. Catheter toronto stentless prosthetic valve implantation. J Heart Valve
Cardiovasc Interv 2005; 66:400–407. Dis 2001; 10:31–34.
38. Ebeid MR, Gaymes CH, McMullan MR, et al. Catheter manage- 57. Moss RR, Ivens E, Pasupati S, et al. Role of Echocardiography in
ment of occluded superior baffle after atrial switch procedures for Percutaneous Aortic Valve Implantation. J Am Coll Cardiol Imag-
transposition of great vessels. Am J Cardiol 2005; 95:782–786. ing 2008; 1:15–24.
39. Sharaf E, Waight DJ, Hijazi ZM. Simultaneous transcatheter occlu- 58. Pavlides GS, Hauser AM, Dudlets PI, et al. The value of trans-
sion of two atrial baffle leaks and stent implantation for SVC esophageal echocardiography in predicting immediate and long-
obstruction in a patient after Mustard repair. Catheter Cardiovasc term outcome of balloon mitral valvuloplasty: comparison with
Interv 2001; 54:72–76. transthoracic echocardiography. J Interv Cardiol 1994; 7:401–408.
40. Crowley DI, Donnelly JP. Use of Amplatzer occlusion devices to 59. Goldstein SA, Campbell A, Mintz GS, et al. Feasibility of on-line
occlude Fontan baffle leaks during fenestration closure proce- transesophageal echocardiography during balloon mitral valvulot-
dures. Catheter Cardiovasc Interv 2008; 71:244–249. omy: experience with 93 patients. J Heart Valve Dis 1994; 3:136–148.
41. Masura J, Bordacova L, Tittel P, et al. Percutaneous management 60. Park SH, Kim MA, Hyon MS. The advantages of On-line trans-
of cyanosis in Fontan patients using Amplatzer occluders. Cath- esophageal echocardiography guide during percutaneous balloon
eter Cardiovasc Interv 2008; 71:843–849. mitral valvuloplasty. J Am Soc Echocardiogr 2000; 13:26–34.
[15/7/010/23:21:31] [526–540]
61. Cortes M, Garcia E, Garcia-Fernandez MA, et al. Usefulness of 63. Shapira Y, Hirsch R, Kornowski R, et al. Percutaneous closure of
transesophageal echocardiography in percutaneous transcatheter perivalvular leaks with Amplatzer occluders: feasibility, safety,
repairs of paravalvular mitral regurgitation. Am J Cardiol 2008; and short-term results. J Heart Valve Dis 2007; 16:305–313.
101:382–386. 64. Silvestry FE, Rodriguez LL, Herrmann HC, et al. Echocardio-
62. Pate GE, Al Zubaidi A, Chandavimol M, et al. Percutaneous graphic guidance and assessment of percutaneous repair for
closure of prosthetic paravalvular leaks: case series and review. mitral regurgitation with the Evalve MitraClip: lessons learned
Catheter Cardiovasc Interv 2006; 68:528–533. from EVEREST I. J Am Soc Echocardiogr 2007; 20:1131–1140.
[2/7/010/10:8:6] [541–553]
42
Percutaneous therapy for aortic valve disease

Peter Wenaweser, Lutz Buellesfeld, and Eberhard Grube
INTRODUCTION AND HISTORICAL PERSPECTIVE approach has been improved constantly over the past years.
Valvular aortic stenosis represents an important cardiac disease The first successful animal studies of porcine valves sutured
affecting up to 5% of the elderly population (1). A severe, into a self-expanding stent were published at the beginning of
symptomatic stenosis signifies a class I indication for valve the new century (19). In 2002, the first-in-man percutaneous
replacement (2), as patients with medical treatment face a high valve implantation within a stenotic aortic valve using the
morbidity and mortality (3). Similar to the pathogenesis of antegrade approach (trans-septal puncture) was achieved by
arteriosclerosis, shear stress, inflammation, and lipid accumu- Cribier et al. (20) using the Cribier-Edwards valve and in the
lation play an important role in the development of aortic following years the Edwards-SAPIEN valve (Edwards Life-
stenosis (4). Attempts to reduce the progression of aortic sciences Inc., California, U.S.). The antegrade approach was
valve disease with different drugs have failed since the use of challenging and prone to complications, and in a second phase,
lipid-lowering drugs (5), or angiotensin-converting enzyme (6) the retrograde approach was used. The retrograde approach
inhibitors have demonstrated no beneficial effect so far. Until turned out to be less demanding and is now the sole approach
2002, the only therapeutic treatment for severe aortic valve used in transfemoral access cases. In 2004 another device, the
stenosis consisted of surgical open-heart valve replacement. CoreValve Revalving System (CRS TM, CoreValve Inc., Irvine,
Perioperative and clinical short-term data show a low opera- California, U.S.) was successfully applied and became the sec-
tion-related morbidity and mortality (7,8), but the incidence of ond technically feasible system (21). As a further step forward,
perioperative, major adverse cardiovascular events increase the first transapical implantation with the Edwards system was
with age (9) and with the number of comorbidities or cardiac introduced into clinical practice in 2005 (22).
factors such as depressed systolic left ventricular function or
concomitant coronary artery disease.
The first interventional treatment of valvular heart dis- ANATOMIC CONSIDERATION
ease goes back to 1953 when Rubio-Alvarez et al. reported on Aortic stenosis represents the most common cause of left
the first intracardiac valvulotomy by means of a catheter (10). ventricular outflow tract obstruction and has three principal
In 1979, “balloon valvulotomy” of a congenital pulmonary etiologies: congenital, rheumatic, and degenerative. The aortic
stenosis was successfully performed (11), and in the following valve can be unicuspid, bicuspid, or triscuspid. A congenitally
decades percutaneous balloon valvuloplasty for pulmonary unicuspid valve is rare and provokes severe symptoms in
valve stenosis become the therapeutic gold standard (12,13). infancy. A congenitally bicuspid aortic valve may degenerate
The technique of balloon valvuloplasty was adopted for the very early in childhood due to turbulent flow inducing trauma
treatment of recoarctation (14), rheumatic mitral valve stenosis to the leaflets and finally resulting in fibrosis and calcification of
(15), and valvular degenerative aortic stenosis (16). After the the valve. Age-related degenerative, former senile, aortic steno-
first promising procedural results of balloon valvuloplasty for sis is the most common cause of acquired aortic stenosis,
valvular aortic stenosis (16), the analysis of larger patient whereas rheumatic disease is rarely pathogenetically involved
populations demonstrated mediocre short-term and poor in Western communities. The major types of aortic stenosis are
long-term clinical outcome (17). Balloon valvuloplasty of heav- depicted in Figure 42.1 (23). The development of degenerative,
ily calcified degenerative aortic stenosis provided an acute gain calcific aortic stenosis shares the risk factors of vascular athero-
of the aortic orifice area by fracturing calcified nodules, sepa- sclerosis. Mechanical stress damages the endothelium of the
rating fused commissures, and stretching of the aortic valve leaflets, facilitating the subendothelial accumulation of oxidized
ring. However, as only microfractures are achieved, the rate of low-density lipoprotein, production of angiotensin II, and
restenosis remains high (*50%) within the first months after inflammation with T lymphocytes and macrophages (4). Pro-
intervention, and the procedure is limited by a considerable gressive calcification leads to severe obstruction and stenosis.
periprocedural morbidity and mortality. As a consequence, A detailed assessment of the aortic valve, aortic root, and
balloon valvuloplasty for the treatment of severe aortic stenosis descending aorta including the iliofemoral axis is warranted
is currently only considered in emergency cases as a bridge to before attempting transcatheter aortic valve implantation
valve replacement (2). (TAVI). First, and of main interest, are the valvular anatomic
In 1992, almost 40 years after the first surgical implanta- details, for estimating the amount and distribution of calcifica-
tion of a prosthetic ball valve device in the descending aorta for tion, as well as the cusps. For balloon dilatation and TAVI, a
the treatment of aortic regurgitation, Anderson et al. succeeded large block of calcification in one cusp and a functionally
in transluminally implanting a prosthetic heart valve in the bicuspid valve represent a formal contraindication for the
ascending aorta of closed chest animals (18). In parallel to the intervention as the risks of emboli, insufficient dilatation, and
improvement of cardiac surgery techniques, the percutaneous a misplacement of the inserted bioprosthesis are markedly
[2/7/010/10:8:6] [541–553]
Figure 42.1 Stenotic aortic valves, operatively excised. (A) Unicuspid, unicommissural valve from a 53-year-old man. (B) Unicuspid,
acommissural valve from a 51-year-old woman. (C) Congenitally bicuspid valve from a 58-year-old man. (D) Tricuspid valve from an 86-year-
old woman. (E) Valve of indeterminate structure (excised in multiple pieces) from a 31-year-old man. Source: From Ref. 53.
increased. Second, the location of the orifices of the coronary SCREENING FOR TAVI
arteries, usually within the two anterior sinuses of Valsalva, The screening exams are an essential condition for a successful
may vary. Of special interest with regard to transcatheter TAVI. Table 42.1 provides the most important exams during the
bioprostheses is the distance between the basal attachment of screening phase categorized into mandatory exams like echo-
the leaflets and the corresponding orifice. In a morphometric cardiography (TTE or TEE), right and left heart catheterization,
and topographic study of the coronary ostia the distance
amounted to 12 to 13 mm (24). The currently used transcatheter
bioprostheses are constructed with a sealed skirt at the distal, Table 42.1 Screening Process for Transcatheter Aortic Valve
left ventricular end of the prosthesis measuring 8 to 11 mm in Implantation
height for the Edwards-SAPIEN valve and 12 mm for the l Mandatory exams
CoreValve Prosthesis. In optimal positioning 4 to 6 mm of the Physical exam
prosthesis will be placed below the annulus providing enough Blood tests
distance of the sealed skirt to the coronary ostia. Dilatation of Electrocardiogram
the aortic root is associated with aortic stenosis, especially with a Transthoracic and transesophageal echocardiography
bicuspid valve (25). The etiology however remains to be deter- Left and right heart catheterization
mined as hemodynamic but also intrinsic pathological factors Aortography (thoracic, abdominal, iliofemoral)
of the aortic wall are involved in the process of dilatation. The Angiography or Doppler ultrasound of carotids
CT angiography (thoracoabdominal)
current guidelines recommend to consider a replacement of the l Ancillary tests
ascending aorta for a diameter >45 mm (2). Accordingly, TAVI Duplex of iliofemoral arteries
is contraindicated in patients with dilated ascending aorta. The Chest X ray
descending aorta and the iliofemoral arteries need to be assessed Pulmonary function test
with regard to technical feasibility of introducing the trans- Neurologic exam, CT scan of brain
catheter valve systems. Severe atherosclerosis of the peripheral Evaluation of renal impairment
vascular bed is often associated with aortic stenosis (26) and Assessment of quality of life
represents a concern for retrograde, transfemoral TAVI with l Estimation of operative risk
respect to risk of dissection, perforation, and embolization of EuroScore
debris. STS Score
[2/7/010/10:8:6] [541–553]
PERCUTANEOUS THERAPY FOR AORTIC VALVE DISEASE 543
Figure 42.3 Cardiac CT scan with contrast. Examples of measur-

ing different diameters of interest during screening for transcatheter
aortic valve implantation: aortic valve annulus (21.6 mm), sinus
Valsalva (29 mm), and ascending aorta (24 mm).
Figure 42.2 CT scan with contrast and 3D reconstruction to eval- measuring of the aortic annulus and the distance between
uate tortuosity of the large vessels. insertion of valve leaflets and the coronary ostia as well as
the dimensions of the sinus Valsalva and the ascending aorta
(Fig. 42.3). Measuring the native aortic annulus can be per-
formed by transthoracic (TTE), transesophageal echocardiog-
and aortography and ancillary tests indicated in certain circum- raphy (TEE), CT scan, or supra-annular aortography. Personal
stances and according to the patient history like, for example, a experience shows that exact measurement is best achieved
neurological exam including a CT scan of the brain in patients using either TEE or CT scan. Of note, the current two devices
with prior stroke (Table 42.1). Echocardiography allows the with CE mark, the CoreValve Revalving System (CRS) and the
exclusion of different contraindications for TAVI-like left ven- Edwards-SAPIEN valve (ESV), cover only about 90% of the
tricular thrombus, severe mitral regurgitation, severe left ven- observed diameters of the aortic annulus. The ESV may be used
tricular hypertrophy, and subvalvular stenosis. A severely for a native annulus measuring between 18 and 25 mm and the
depressed left ventricular systolic function represents a high- CRS for an annulus diameter of 19 to 27 mm. A patient selection
risk condition and indicates the need for a hemodynamic assist matrix provided by the companies was introduced in clinical
(e.g., TandemHeart or ECMO) during the procedure or a practice and facilitate the screening of candidates (Fig. 42.4).
contraindication for an intervention. The invasive evaluation Apart from the screening of technical feasibility, the
embodies the most important part of the screening assessing clinical part often turns out to be more challenging for the
valve area, cardiac output, pulmonary artery pressure, and treating physician as the intended patient population is old of
pulmonary and systemic vascular resistance by hemodynamic age and at high-risk for operative complications. Last but not
measurements. Furthermore, the retrograde passage of the least the general clinical condition and the accompanying
calcified valve is attempted, and concomitant coronary artery frailty of the patient may turn the scale to turn down the
disease may be treated with percutaneous coronary interven- candidate. Geriatric assessment tools need to be tested to better
tion during the screening procedure. Aortography provides a identify patients at high risk for postprocedural perturbation
good image of the annulus-to-aorta angle and the angle of the and prolonged rehabilitation.
aortic arch as well as the calcification of the large arteries. CT
scan may further help to estimate the angulation of the annu-
lus-to-aorta part and the degree of calcification of the valve, the PATIENT SELECTION AND RISK ASSESSMENT
aorta, and the peripheral vessels. Additional use of contrast The patient selection represents the most challenging part of
identifies the lumen and tortuosity of the aorta (Fig. 42.2) and the whole process from screening until TAVI. Risk assessment
the great arteries and is nowadays mostly performed. It may plays a crucial role and should consist of multidisciplinary
preclude problems during the introduction of the devices. consultation between cardiologists, cardiac surgeons, anes-
Detailed imaging of the route of implantation during catheter- thesiologists, and other specialists previously involved in the
ization may allow to omit the CT scan in certain condition but treatment of the patient. The evaluation of the risk for TAVI is
latter is currently mostly performed as a two-dimensional based on the estimation of the risk for conventional surgery.
reconstruction of the aortic valve and root allows an exact Different risk scores have been created and validated. The
[2/7/010/10:8:6] [541–553]
Figure 42.4 Patient selection matrix for the CoreValve Revalving System. Source: Courtesy of CoreValve Inc., Irvine, California, U.S.
EuroScore (27), the STS-predicted risk mortality score (28), and Table 42.2 Indications for TAVI.
the Ambler score (29) are among others of main interest. l Severe symptomatic aortic stenosis (aortic valve area <1 cm2 or
However, several limitations of these scores need to be kept <0.6 cm2/m2 or a mean gradient >40 mmHg) and
in mind. The predictive value in high-risk patients is limited, as l Contraindication for conventional surgery (e.g., previous chest
especially the EuroScore tends to overestimate the risk of mor- radiation, porcelain aorta)
tality in patients with high-risk features (30). As high-risk or
patients represent a small part of the patient population l Logistic EuroScore >20% or STS Score >10%
included into the models for generating a score, the predictive Source: From Ref. 32.
ability of the scores is limited, last but not least because of a
large heterogeneity of the patient characteristics.
The primary indication for TAVI were patients with at high-risk or with contraindications for surgery. They under-
severe (aortic valve area <1 cm2 or <0.6 cm2/m2), symptomatic scored the point that it is too premature to establish TAVI in
aortic stenosis who were refused for conventional surgery. patients who are good surgical candidates (32). The estimation
Correspondingly, the initial studies with the ESV and the of risk for surgery is based, therefore, not only on quantitative
CRS included patients with contraindications for conventional assessment of risk scores but also on clinical judgment. The
open-heart surgery, like porcelain aorta, but also high-risk formal indication criteria are depicted in Table 42.2.
patients for surgery with a logistic EuroScore 20% or an STS
Score 10%. For the 18 Fr CRS study patients with age
80 years irrespective of the risk calculation (31). After publi- DEVICES AND TECHNIQUE
cation of the first clinical results showing procedural success Edwards-SAPIEN Valve
rates of approximately 90% and after CE approval of the two The Edwards-SAPIEN prosthesis (Edwards Lifesciences)
devices, the European Association of Cardio-Thoracic Surgery (Fig. 42.5) is based on the prototype PVT prosthesis (Percuta-
(EACTS) and the European Society of Cardiology (ESC) in neous Valve Technologies, Inc.) first described and clinically
collaboration with the European Association of Percutaneous tested by Alain Cribier et al. in 2002 (20). The ESV consists of a
Cardiovascular Interventions (EAPCI) published in a position tubular slotted stainless steel stent with an attached bovine
statement that TAVI should currently be restricted to patients pericardial trileaflet valve and fabric-sealing cuff. Two sizes are
[2/7/010/10:8:6] [541–553]
Figure 42.5 Edwards-SAPIEN valve (Edwards Lifesciences Inc.,

California, U.S.).
Figure 42.7 CoreValve Revalving System (CoreValve Inc., Irvine,

California, U.S.).
Figure 42.6 Steerable Retroflex catheter for delivery of the

Edwards-SAPIEN valve.
available: the 23- and the 26-mm device. Immediately before

implantation, the prosthetic stent valve is mechanically Figure 42.8 Comparison of three CoreValve device generations:
25-Fr delivery catheter (top), 21-Fr delivery catheter (middle), and
crimped onto a balloon catheter. The smaller prosthesis is
18-Fr delivery catheter (bottom).
intended to be expanded with a 22-mm diameter balloon to
achieve a 23-mm external diameter; the larger prosthesis is
intended to be expanded with a 25-mm balloon to achieve a
26-mm external diameter. The 23-mm valve is considered suit-
able for an annulus diameter of 18 to 22 mm and the 26 mm
exclude the calcified leaflets and to avoid recoil. The middle
valve for an annulus diameter of 21 to 25 mm. In case of
portion is constrained to avoid the coronary arteries, while the
transfemoral access, a 22-Fr (23 mm device) or a 24-Fr (26 mm
upper portion is flared to center and fix the stent frame in the
device) introducer sheath is needed. The prosthesis is then
ascending aorta and to provide longitudinal stability and coax-
delivered using a steerable catheter, the FlexCath device
ial positioning. The pericardial valve is located in the transmis-
(Fig. 42.6). In case of transapical approach, a 33-Fr introducer
sion zone between the lower and the middle part, resulting in a
sheath is being used.
supra-annular position compared to the native valve. The sizes
of three subsequently developed delivery systems have been
CoreValve Revalving Prosthesis gradually reduced from 25 to 18 Fr over time to facilitate
The CoreValve Revalving prosthesis (CoreValve), first vascular access and deployment of the device (Figs. 42.8
described by Eberhard Grube et al. 2006 (21), consists of a and 42.9). The currently commercially available third-genera-
trileaflet bioprosthetic porcine pericardial tissue valve, which is tion device (18F) is offered in two different sizes for different
mounted and sutured in a self-expanding nitinol stent frame annulus dimensions: the small 26 mm prosthesis for aortic
(Fig. 42.7). The prosthetic frame is manufactured by laser cut- valve annulus sizes from 20 to 24 mm and the large 29 mm
ting of a nitinol metal tube with a length of 50 mm. The lower prosthesis for aortic valve annulus sizes from 24 to 27 mm. A
portion of the prosthesis has high radial force to expand and comparison of device characteristics is depicted in Table 42.3.
[2/7/010/10:8:6] [541–553]
Rapid pacing by a provisional pacemaker in the right ventricle

helps avoiding a dislocation of the balloon into the ventricle
and minimizes the risk of wire-induced perforation of the
ventricle. After dilatation of the native valve, the bioprosthesis
is inserted, directed around the aortic arch (using the Retroflex
catheter with the Edwards system) and deployed under fluo-
roscopic guidance and repeated dye injections. An effective
rapid pacing with drop of systolic blood pressure down to
40 mmHg is only mandatory for the deployment of the ESV (33)
but not for the CRS. Functional assessment of the bioprosthesis
is assessed by simultaneous measurement of pressures in the
Figure 42.9 18-Fr CoreValve delivery catheter consisting of the ascending aorta and the left ventricle. Aortic regurgitation is
distal housing, the shaft, and the handle for valve release. evaluated primarily by aortography and eventually further
assessed by TEE to discriminate between transvalvular and
paravalvular regurgitation.
Table 42.3 Comparison of Edwards-SAPIEN and CoreValve Routes of Implantation

Device Characteristics Antegrade Approaches
The transfemoral antegrade approach using a femoral vein with
Edwards-SAPIEN CoreValve
trans-septal puncture has been primarily used in the early
Stent material Stainless steel Nitinol beginning of transcatheter aortic valve replacement (20). After
Valve material Bovine pericardium Porcine pericardium standard trans-septal catheterization, a straight 0.035-in. guide-
Release mechanism Balloon expansion Self-expandable
wire was advanced across the stenotic aortic valve through a
Diameter of valve 18–25 mm 19–27 mm
annulus balloon flotation catheter. After advancement of the balloon
Sheath size 22, 24 Fr 18 Fr catheter into the descending aorta, the guidewire was
Femoral artery 8 mm 6 mm exchanged for a stiff long guidewire, which was snared from
diameter the left femoral arterial access site to create a loop for optimal
Valve size 23, 26 mm 23, 27 mm device deployment support when advancing the device ante-
Worldwide number >3500 >3500 gradely. Additional septal dilation was usually necessary to
of implantations bring the prosthesis through the intra-atrial septum. Because of
Transapical Yes No the complexity as well as the higher complication rate com-
application pared to the retrograde access, the transfemoral antegrade
Transaxillary No Yes
approach has been almost completely abandoned. Potential
application
CE certification Yes Yes interferences with the mitral valve apparatus during the pro-
FDA approval No No cedure were some of the specific severe complications leading
to hemodynamic instability in some cases reported by Cribier
et al. (34).
The transapical antegrade approach is a popular surgical
approach for transcatheter aortic valve implantations. This tech-
Procedure nique has been described first for the balloon-expandable valve
TAVI is performed either under local anesthesia and mild prosthesis (Cribier-Edwards) in 2006 by Lichtenstein et al. (22).
sedation in case of a pure transcatheter, retrograde approach The technique requires an anterolateral minithoracotomy, peri-
using suture devices or under general anesthesia in case of a cardiotomy, and then puncture of the apex of the left ventricle
transapical approach or surgical cut-down. Prophylactic use of using a needle through purse-string sutures. Today, this
antibiotics and the use of unfractionated heparin (targeted technique is evolving to a new standard surgical procedure par-
ACT: 200–250 seconds) are standard of care. The access for ticularly for patients considered for catheter-based valve implan-
the large sheath is mostly gained from the right groin as the tation but unfavorable peripheral arteries. However, the usage of
right iliofemoral axis is often less tortuous. From the left com- transapical versus truly percutaneous transfemoral approaches is
mune femoral artery a pigtail is advanced to the descending significantly varying among heart centers, depending on operator
aorta followed by angiographic visualization of the right iliofe- experiences and local politics. Being more invasive than a trans-
moral axis before puncture. In the following, the pigtail is used femoral approach, the transapical access is a valuable route for
to mark the level of the native annulus. Repeated injections of patients with inappropriate peripheral access. Cardiopulmonary
dye help to evaluate the degree of aortic regurgitation during bypass should be on standby in patients in whom surgical con-
and after the deployment of the prosthesis. After introducing a version is an option in case of complications.
closure device and a 9-Fr sheath over a regular wire, the native, Given the particular deployment characteristics, only the
stenotic valve is usually passed retrogradely with a straight Edwards balloon-expandable valve prosthesis has currently the
wire directed by an Amplatz left 1 catheter. An extra-stiff wire option for antegrade insertion.
(e.g., Amplatz extra or superstiff short tip), preshaped manu-
ally like a pigtail, is then carefully placed into the left ventricle. Retrograde Approaches
The large 18-Fr (CoreValve) or 22-Fr/24-Fr sheath (Edwards) is The retrograde approach is presently the most popular way for
then inserted over the stiff wire followed by the balloon TAVI. Various arterial access sites are iliac, femoral, or sub-
valvuloplasty (e.g., with a Nucleus or a Tyshak II balloon). clavian access.
[2/7/010/10:8:6] [541–553]
Table 42.4 Advantages and Disadvantages of Antegrade and Retrograde Approaches

Transfemoral antegrade Transapical antegrade Transfemoral retrograde
Advantages l No large arterial access needed, l Direct access with antegrade l Ease of technique
less arterial complications crossing l No interference with other cardiac
l No need for surgical preparations l Independent of peripheral structures as opposed to the
l Antegrade crossing of the aortic vessel status transfemoral antegrade technique
valve ¼> potentially less prone
to cause embolizations
Disadvantages l Technically more challenging l Invasive surgical technique l Large arterial access needed
(longer route, trans-septal l Requires general anesthesia l Retrograde aortic valve crossing
puncture)
l Interference with mitral valve
¼>; leads to.
First transcatheter retrograde device implantations high volume centers. Given the initial study enrollment criteria
described by Webb et al. (Cribier-Edwards prosthesis) (35) as well as the present conventional treatment practice recom-
and our group (CRS) (36) have been performed via the iliac mending surgery as the gold standard, patients are mostly
access given the profile size of the early device prototypes. above 80 years old and considered high risk or with contra-
After surgical cut-down and subsequent wire placement into indications for surgical valve replacement.
the left ventricle, the devices are advanced retrogradely
through the native aortic valve and then implanted. Retrograde
crossing of heavily calcified aortic valves may occasionally be Procedural Success and Clinical Outcome
problematic, but technical developments with lower device The procedural success rate increased remarkably over the past
profiles as well as steerable catheters facilitated this procedural two to four years with rates between 90% and 99% in experi-
part remarkably. The progressive device size reduction over the enced hands (Tables 42.5 and 42.6) (38) Mortality at 30-day
past five years led to the truly percutaneous access in the follow-up is between 6% and 15%, procedural myocardial
common femoral artery. Reliable access site closure is usually infarction rate 0.7% and 16%, procedural stroke rate 1.4% and
achievable using a Prostar1 XL suture device (Abbott Vascular, 9% with current devices (Tables 42.5 and 42.6). Direct coronary
Abbott Park, Illinois, U.S.) or multiple “closure” suture devices obstruction is rare (<1%) and is only reported for the Edwards
that are preloaded at the beginning of the procedure. Fluoro- valve. Permanent pacemakers due to significant postprocedural
scopic guidance should be mandatory for correct puncture; atrioventricular conduction disturbances are needed in up to
echo guidance might be helpful as well. 20% of patients (39,40). It is currently unknown if results are
The subclavian access has been first described by our affected by the device itself, but according to the first reports,
group as an alternative route for patients with diseased iliac/ periprocedural myocardial infarction—although relatively
femoral arteries (31,36,37). Access can be obtained after surgical rare—is observed more frequently with the ESV than with the
cut-down either directly or after placement of a graft trunk onto
the subclavian artery to both avoid access dissections and
facilitate final closure. Because of the higher technical complex- Table 42.5 Edwards Short-Term Clinical Outcome
ity as well as the need for surgical preparations, this approach REVIVE II REVIVAL II
should be considered a backup technique in case that standard (N ¼ 52) (N ¼ 124)
alternative insertion sites are inaccessible. Procedural success rate (%) 88.0 87.3
The direct surgical aortic access with thoracotomy and 30-Day all mortality (%) 13.2 7.3
retrograde catheter-based implantation of a valve prosthesis Myocardial infarction (%) 8.5 16.3
is a new technique just recently tested by cardiac surgeons. As Neurologic events (%) 2.8 9.0
compared to standard surgical aortic valve replacements, there Source: From Ref. 51.
is no need of suture fixation of the prosthesis when using one of
the catheter-based balloon-expandable or self-expanding devi-
ces. However, this technique is still in its infancy. Table 42.6 CoreValve Short-Term Clinical Outcome
The advantages and disadvantages of antegrade and
retrograde approaches are depicted in Table 42.4. 21 Fr 18 Fr 18 Fr
S&E study S&E study EE registry
(N ¼ 52) (N ¼ 124) (N ¼ 1243)
CLINICAL ASPECTS AND OUTCOME Procedural success 90.4 94.4 98.2

rate (%)
The technology of transcatheter aortic valve implantation has
30-Day all mortality (%) 15.4 14.5 6.7
significantly improved with the development of delivery cath- Cardiac deaths (%) 7.7 11.2 3.9
eters with smaller profiles and better prostheses with various Myocardial infarction 3.8 3.4 0.7
size options. Together with increasing operator experience, (%)
these progresses are reflected in the improved clinical outcome Pacemaker (%) 17.3 25.8 12.2
observed over the past few years. Today, more than 4000 Stroke (%) 17.3 6.5 1.4
patients have been treated with one of these two CE-approved TIA (%) 0.0 5.6 0.3
devices with almost exponentially increasing numbers partic- Abbreviations: S&E, Safety and efficacy evaluation; EE, extended
ularly in the last 18 months. However, scientific outcome evaluation.
reports are still limited, mainly originating from few leading Source: From Ref. 52.
[2/7/010/10:8:6] [541–553]
Figure 42.10 Heart failure NYHA class among patients (N ¼ 152/

107/67/13) undergoing transcatheter aortic valve replacement with
Edwards-SAPIAN in the Revive II + Revival II studies. Source: From
Ref. 51.
Figure 42.12 Long-term survival after CoreValve implantation.
Figure 42.11 Heart failure NYHA class among patients (N ¼ 134)

undergoing transcatheter aortic valve replacement with the Core-
Valve Revalving System. Source: From Ref. 52. Figure 42.13 Long-term survival after transapical aortic valve
replacement with Edwards. Source: From Ref. 54.
CRS. On the other hand, the CRS is associated with a larger

need for permanent pacemaker implantation. To address these
reduction of the mean transvalvular gradient to around
adverse effects in detail, more data and eventually randomized
10 mmHg with a mean neo-orifice area of 1.5 to 1.8 cm2 (Figs.
trials are warranted.
42.14 and 42.15). Paravalvular aortic regurgitation, mostly mild
Patients with successful implantation procedures and
or mild to moderate, is observed in around 70% of patients and
uneventful 30-day follow-up improve clinically and benefit
partially related to focal excessive load of calcifications
durably from the procedure as expressed in the NYHA grade
(Figs. 42.16 and 42.17) that prohibits optimal device expansion
comparisons (Figs. 42.10 and 42.11) as well as the survival
with complete paravalvular sealing. However, severe regurgita-
curves (Figs. 42.12 and 42.13). Immediate or late embolization
tions are rare given the improved screening process (matching
of a valve prosthesis is also rare, reported only for the Edwards
dimensions of annulus and prosthesis) as well as various proce-
device (41).
dural techniques and strategies to deal with high-degree regur-
gitations occurring immediately after device deployment. In
Echocardiographic Outcome long-term follow-up, serial echocardiographic studies have con-
As seen in all published series, the hemodynamic status improves sistently shown good prosthetic valve function with no structural
immediately after successful device implantations with a deterioration of valve tissue (38,42,43).
[2/7/010/10:8:6] [541–553]
Figure 42.14 Echocardiographic outcome, mean pressure gradient, pooled Revive II + Revival II, Edwards. Source: From Ref. 51.
Figure 42.15 Echocardiographic outcome, pressure gradients, CoreValve. Source: From Ref. 52.
Outcome for Transapical Technique Mortality rates range from 9% to 18%. Strokes occur in about
More than 400 patients with significant aortic valve stenosis 0% to 6%. Nonrandomized comparison of transapical and
have presently been treated by transapical aortic valve implan- transfemoral approaches from the SOURCE and the PARTNER
tation. Reports are available only for the ESV from two expe- EU trial indicate a higher mortality rate for the transapical
rienced centers (22,44,45). Further data of the TRAVERCE approach in selective patients with higher estimated risk than
study (Fig. 42.18), the SOURCE registry, and the PARTNER patients undergoing retrograde TAVI. Randomized studies are
EU registry were presented at TCT 2008. Implantation success needed to finally assess superiority of one technique over the
rate was about 90%. The majority of patients is being done off other.
pump, and the rate of perioperative conversion is 9% to 12%.
[2/7/010/10:8:6] [541–553]
Figure 42.16 Echocardiographic outcome, aortic regurgitation, pooled Revive II + Revival II, Edwards. Source: From Ref. 51.
Figure 42.17 Echocardiographic outcome, aortic regurgitation, CoreValve. Source: From Ref. 52.
LIMITATIONS embodies another potential target for the use of transcath-

Several limitations with regard to percutaneous treatment of eter bioprosthesis. However, an only mildly calcified native
aortic stenosis need to be addressed. valve may hinder a sufficient anchoring of the prosthesis
l
and moreover the often-associated dilatation of the ascend-
First, the simple approach of balloon valvuloplasty has
ing aorta might contraindicate the sole treatment of the
failed showing significant improvement of clinical symp-
valve problem. Nevertheless, in patients with prior surgical
toms and increasing the valve area in the midterm follow-
valve replacement (biological valve), TAVI appears to
up. In addition, a considerable rate of complications were
emerge as valuable treatment option.
associated with the procedure leading to abandoning this l Third, vascular access limits the use of transcatheter-based
method (17,46).
l
devices in a substantial number of patients. Further tech-
Second, degenerative, former senile, aortic stenosis repre-
nical improvements are required to provide smaller
sents currently the only considerable indication of TAVI in a
sheaths and to reduce the associated risk of peripheral
selective, high-risk patient population. Aortic insufficiency
[2/7/010/10:8:6] [541–553]
and operator experience markedly reduces this type of

complication. Cardiac tamponade provoked by wire per-
foration is rare (*1%) and associated with the presently
inevitable use of a stiff wire for the placement of the
currently used devices. Patients with prior large myocar-
dial infarction or chronic steroid treatment maybe at higher
risk for this complication.
SPECIAL ISSUES AND FUTURE DEVELOPMENTS

There are multiple new concepts under scientific as well as
clinical evaluation designed to provide new features overcom-
ing some of the shortcomings of the first-generation techniques.
Only two already clinically tested devices are mentioned in the
following:
The SADRA Lotus valve, for example, is a self-expanding
device providing the ability of correcting and repositioning the
valve prosthesis before releasing and permanently seating the
device. The user is able to verify functionality and anchoring
properties as well as the position of the device in relationship to
the coronary arteries, the native aortic valve, and the mitral
Figure 42.18 Long-term survival in the TRAVERCE study. Source: apparatus before final implantation (48) (Fig. 42.19). As
From Ref. 51.
opposed to the “one-shot” release of the balloon-expandable
techniques, this technology of repositioning aims to offer a
more controlled way of device deployment. Once the device is
in correct position, a locking mechanism stabilizes the system.
In addition, to avoid a paravalvular leak, a sealing membrane is
vascular complications. Latter have been reported with an
attached to the outside of the prosthesis filling out paravalvular
incidence ranging from 2% to 15% remaining a substantial
gaps. The fist-in-man study using the SADRA Lotus valve is
cause of morbidity and mortality (31,34–36,38,40,47). Recent
currently ongoing.
data reveal however that an operator learning curve and the
The DirectFlow valve is another new concept that is
use of smaller sheaths as well as a more detailed screening
repositionable and retrievable at all time points during the
help reducing this type of complication (38).
l
procedure (49). The device consists of a stentless tissue valve
Forth, myocardial infarction is rarely associated with TAVI
with bovine pericardial leaflets, which are located within an
and occurs only in cases of misplacement of the valve.
l
expandable nonmetallic framework. Results of the presently
Fifth, major cerebrovascular events occur in 2% to 4%
ongoing first-in-man study are shortly available.
(40,43) and maybe related with balloon dilatation, passing
The indication for TAVI is currently limited to a selective,
of the aortic arch with the device or a hemodynamic
high-risk patient population with severe, symptomatic aortic
compromise during valve placement. Indirect comparison
stenosis but further possible indications have already been
with surgical data shown comparable event rates (9), but
tested clinically or by compassionate use. The use of TAVI for
further technical developments need to address this pro-
the indication of a stenotic or insufficient degenerated bioprosthesis
cedure-related limitation of TAVI.
l
appears very attractive. Two major arguments favor TAVI over
Sixth, mild-moderate aortic regurgitation is observed in up
conventional cardiac reoperation. Latter is complicated by rest-
to 50% of the treated patient populations, but severe,
ernotomy and the positioning of the new valve is facilitated by
clinically relevant insufficiency remains rare in case of
clear visualization of the annulus. The CRS (50) as well as the
correct matching of the prosthesis with the size of the
aortic annulus as well as in case of correct positioning.
l Seventh, the need of a permanent pacemaker implantation
after TAVI is reported in up to 20% of the cases. CRS seems
to induce more frequently atrioventricular disturbances,
probably due to the sustained radial force of the Nitinol
frame and an interference with the subvalvular, closely
situated conduction system.
l Eighth, embolization of the device is very rare (<1%). It is
related either with a mismatch of the diameters (device—
annulus) or with a misplacement of the bioprosthesis dur-
ing deployment. The CRS provides the possibility retriev-
ing the not completely deployed valve into the sheath.
After reloading into the delivery catheter, the same bio-
prosthesis can be positioned again.
l Ninth, severe bleeding is caused either by vascular access
complication or by vessel dissection/perforation induced Figure 42.19 Lotus Valve (Sadra Medical, Inc., California, U.S.).
by the sheaths. Careful screening of implantation routes
[2/7/010/10:8:6] [541–553]
REFERENCES
1. Nkomo VT, Gardin JM, Skelton TN, et al. Burden of valvular heart
diseases: a population-based study. Lancet 2006; 368:1005–1011.
2. Bonow RO, Carabello BA, Chatterjee K, et al. ACC/AHA 2006
guidelines for the management of patients with valvular heart
disease: a report of the American College of Cardiology/American
Heart Association Task Force on Practice Guidelines (writing
committee to revise the 1998 Guidelines for the Management of
Patients With Valvular Heart Disease): developed in collaboration
with the Society of Cardiovascular Anesthesiologists: endorsed
by the Society for Cardiovascular Angiography and Interventions
and the Society of Thoracic Surgeons. Circulation 2006; 114:
e84–e231.
3. Varadarajan P, Kapoor N, Bansal RC, et al. Clinical profile and
natural history of 453 nonsurgically managed patients with severe
aortic stenosis. Ann Thorac Surg 2006; 82:2111–2115.
4. Otto CM. Calcific aortic stenosis—time to look more closely at the
valve. N Engl J Med 2008; 359:1395–1398.
5. Rossebo AB, Pedersen TR, Boman K, et al. Intensive lipid lowering
with simvastatin and ezetimibe in aortic stenosis. N Engl J Med
Figure 42.20 Final result after transapical Edwards-SAPIEN valve 2008; 359:1343–1356.
implantation in a patient with severe regurgitation after surgical 6. Rosenhek R, Rader F, Loho N, et al. Statins but not angiotensin-
valve replacement with a bioprosthesis. (black arrow) Ring of sur- converting enzyme inhibitors delay progression of aortic stenosis.
gically implanted bioprosthesis (mitroflow 23 mm); (white arrow) Circulation 2004; 110:1291–1295.
Edwards-SAPIEN valve (“crown shaped”) 23 mm. 7. Astor BC, Kaczmarek RG, Hefflin B, et al. Mortality after aortic
valve replacement: results from a nationally representative data-
base. Ann Thorac Surg 2000; 70:1939–1945.
8. Tabata M, Umakanthan R, Cohn LH, et al. Early and late outcomes
of 1000 minimally invasive aortic valve operations. Eur J Cardio-
thorac Surg 2008; 33:537–541.
ESV have already been successfully used for the treatment of 9. Alexander KP, Anstrom KJ, Muhlbaier LH, et al. Outcomes of
degenerated bioprosthesis with severe regurgitation (Fig. 42.20). cardiac surgery in patients > or ¼ 80 years: results from the
The use of TAVI for this indication is limited by the inner National Cardiovascular Network. J Am Coll Cardiol 2000; 35:
diameter of the previously implanted valve and a potential 731–738.
infectious etiology of the degeneration of the valve that needs 10. Rubio-Alvarez V, Limon R, Soni J. Intracardiac valvulotomy by
to be ruled out before intervention. Aortic regurgitation repre- means of a catheter.. Arch Inst Cardiol Mex 1953; 23:183–192.
sents another possible indication for TAVI but is limited by the 11. Semb BK, Tjonneland S, Stake G, et al. “Balloon valvulotomy” of
congenital pulmonary valve stenosis with tricuspid valve insuffi-
usually mild calcification of the valve and the often associated
ciency. Cardiovasc Radiol 1979; 2:239–241.
dilated ascending aorta. 12. Chen CR, Cheng TO, Huang T, et al. Percutaneous balloon
valvuloplasty for pulmonic stenosis in adolescents and adults. N
Engl J Med 1996; 335:21–25.
CONCLUSIONS 13. Kan JS, White RI Jr., Mitchell SE, et al. Percutaneous balloon
Transcatheter aortic valve replacement represents an innova- valvuloplasty: a new method for treating congenital pulmonary-
tive, rapidly evolving technique with already proven feasibility valve stenosis. N Engl J Med 1982; 307:540–542.
in a selective patient population. The high technical success rate 14. Lababidi Z. Aortic balloon valvuloplasty. Am Heart J 1983;
up to 99% is related to an improved screening process, 106:751–752.
advanced device properties like a reduced profile size in com- 15. Inoue K, Owaki T, Nakamura T, et al. Clinical application of
parison with the initial generation as well as improved operator transvenous mitral commissurotomy by a new balloon catheter.
skills and the interdisciplinary team. In experienced hands, the J Thorac Cardiovasc Surg 1984; 87:394–402.
16. Cribier A, Savin T, Saoudi N, et al. Percutaneous transluminal
procedure may be performed under local anesthesia, in less
valvuloplasty of acquired aortic stenosis in elderly patients: an
than an hour without hemodynamic or surgical support. Fur-
alternative to valve replacement? Lancet 1986; 1:63–67.
thermore, the present experience demonstrates the large poten- 17. Otto CM, Mickel MC, Kennedy JW, et al. Three-year outcome after
tial of this technique currently revolutionizing the treatment balloon aortic valvuloplasty. Insights into prognosis of valvular
practice for patients with significant aortic valve disease. aortic stenosis. Circulation 1994; 89:642–650.
Today, surgical valve replacement is still standard of care, 18. Andersen HR, Knudsen LL, Hasenkam JM. Transluminal implan-
but less invasive transcatheter methods are prone to challenge tation of artificial heart valves. Description of a new expandable
this standard in the future. A prerequisite is, however, that the aortic valve and initial results with implantation by catheter
encouraging clinical results are confirmed in larger patient technique in closed chest pigs. Eur Heart J 1992; 13:704–708.
populations, and further progress of the technology is 19. Lutter G, Kuklinski D, Berg G, et al. Percutaneous aortic valve
replacement: an experimental study. I. Studies on implantation.
achieved. Moreover, valve durability needs to be clinically
J Thorac Cardiovasc Surg 2002; 123:768–776.
proven and further achievements with respect to in positioning
20. Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous trans-
and repositioning as well as with regard to embolic protection catheter implantation of an aortic valve prosthesis for calcific
are warranted to improve the procedural safety. Last but not aortic stenosis: first human case description. Circulation 2002;
least, head-to-head comparisons between surgery and TAVI in 106:3006–3008.
different patient subsets will help us to discriminate the best 21. Grube E, Laborde JC, Zickmann B, et al. First report on a human
therapeutic option in the future. percutaneous transluminal implantation of a self-expanding valve
[2/7/010/10:8:6] [541–553]
prosthesis for interventional treatment of aortic valve stenosis. 38. Webb JG, Pasupati S, Humphries K, et al. Percutaneous trans-
Catheter Cardiovasc Interv 2005; 66:465–469. arterial aortic valve replacement in selected high-risk patients
22. Lichtenstein SV, Cheung A, Ye J, et al. Transapical transcatheter with aortic stenosis. Circulation 2007; 116:755–763.
aortic valve implantation in humans: initial clinical experience. 39. Bleiziffer S, Ruge H, Mazzitelli D, et al. Results of percutaneous
Circulation 2006; 114:591–596. and transapical transcatheter aortic valve implantation performed
23. Roberts WC, Ko JM. Frequency by decades of unicuspid, bicuspid, by a surgical team. Eur J Cardiothorac Surg 2009.
and tricuspid aortic valves in adults having isolated aortic valve 40. Piazza N, Grube E, Gerckens U, et al. Procedural and 30-day
replacement for aortic stenosis, with or without associated aortic outcomes following transcatheter aortic valve implantation using
regurgitation. Circulation 2005; 111:920–925. the third generation (18 Fr) corevalve revalving system: results
24. Cavalcanti JS, de Melo NC, de Vasconcelos RS. Morphometric and from the multicentre, expanded evaluation registry 1-year follow-
topographic study of coronary ostia. Arq Bras Cardiol 2003; ing CE mark approval. EuroIntervention 2008; 4:242–249.
81:359–362, 5–8. 41. Clavel MA, Dumont E, Pibarot P, et al. Severe valvular regurgi-
25. Wilton E, Jahangiri M. Post-stenotic aortic dilatation. J Cardio- tation and late prosthesis embolization after percutaneous aortic
thorac Surg 2006; 1:7. valve implantation. Ann Thorac Surg 2009; 87:618–621.
26. Aronow WS, Ahn C, Kronzon I. Association of valvular aortic 42. Ruiz CE, Laborde JC, Condado JF, et al. First percutaneous trans-
stenosis with symptomatic peripheral arterial disease in older catheter aortic valve-in-valve implant with three year follow-up.
persons. Am J Cardiol 2001; 88:1046–1047. Catheter Cardiovasc Interv 2008; 72:143–148.
27. Roques F, Nashef SA, Michel P, et al. Risk factors and outcome in 43. Ye J, Cheung A, Lichtenstein SV, et al. Transapical transcatheter
European cardiac surgery: analysis of the EuroSCORE multina- aortic valve implantation: 1-year outcome in 26 patients. J Thorac
tional database of 19030 patients. Eur J Cardiothorac Surg 1999; Cardiovasc Surg 2009; 137:167–173.
15:816–822; discussion 22–23. 44. Walther T, Simon P, Dewey T, et al. Transapical minimally
28. Clark RE. Calculating risk and outcome: The Society of Thoracic invasive aortic valve implantation: multicenter experience. Circu-
Surgeons database. Ann Thorac Surg 1996; 62:S2–S5; discussion lation 2007; 116:I240–I245.
S31–32. 45. Ye J, Cheung A, Lichtenstein SV, et al. Six-month outcome of
29. Ambler G, Omar RZ, Royston P, et al. Generic, simple risk transapical transcatheter aortic valve implantation in the initial
stratification model for heart valve surgery. Circulation 2005; seven patients. Eur J Cardiothorac Surg 2007; 31:16–21.
112:224–231. 46. Lieberman EB, Bashore TM, Hermiller JB, et al. Balloon aortic
30. Gummert JF, Funkat A, Osswald B, et al. EuroSCORE overesti- valvuloplasty in adults: failure of procedure to improve long-term
mates the risk of cardiac surgery: results from the national registry survival. J Am Coll Cardiol 1995; 26:1522–1528.
of the German Society of Thoracic and Cardiovascular Surgery. 47. Cribier A, Eltchaninoff H, Tron C, et al. Early experience with
Clin Res Cardiol 2009. percutaneous transcatheter implantation of heart valve prosthesis
31. Grube E, Schuler G, Buellesfeld L, et al. Percutaneous aortic valve for the treatment of end-stage inoperable patients with calcific
replacement for severe aortic stenosis in high-risk patients using aortic stenosis. J Am Coll Cardiol 2004; 43:698–703.
the second- and current third-generation self-expanding Core- 48. Buellesfeld L, Gerckens U, Grube E. Percutaneous implantation
Valve prosthesis: device success and 30-day clinical outcome. of the first repositionable aortic valve prosthesis in a patient
J Am Coll Cardiol 2007; 50:69–76. with severe aortic stenosis. Catheter Cardiovasc Interv 2008;
32. Vahanian A, Alfieri OR, Al-Attar N, et al. Transcatheter valve 71:579–584.
implantation for patients with aortic stenosis: a position statement 49. Low RI, Bolling SF, Yeo KK, et al. Direct flow medical percuta-
from the European Association of Cardio-Thoracic Surgery neous aortic valve: proof of concept. EuroIntervention 2008;
(EACTS) and the European Society of Cardiology (ESC), in collab- 4:256–261.
oration with the European Association of Percutaneous Cardiovas- 50. Wenaweser P, Buellesfeld L, Gerckens U, et al. Percutaneous
cular Interventions (EAPCI). Eur J Cardiothorac Surg 2008; 34:1–8. aortic valve replacement for severe aortic regurgitation in degen-
33. Webb JG, Pasupati S, Achtem L, et al. Rapid pacing to facilitate erated bioprosthesis: the first valve in valve procedure using the
transcatheter prosthetic heart valve implantation. Catheter Cardi- corevalve revalving system. Catheter Cardiovasc Interv 2007;
ovasc Interv 2006; 68:199–204. 70:760–764.
34. Cribier A, Eltchaninoff H, Tron C, et al. Treatment of calcific aortic 51. Leon M. Endovascular Aortic Valve Replacement: Experience
stenosis with the percutaneous heart valve: mid-term follow-up with a Balloon Expandable Device. In: SOLACI 2008. Cancun,
from the initial feasibility studies: the French experience. J Am Mexico; 2008.
Coll Cardiol 2006; 47:1214–1223. 52. Grube E, Buellesfeld L, Mueller R, et al. Progress and current
35. Webb JG, Chandavimol M, Thompson CR, et al. Percutaneous status of percutaneous aortic valve replacement: results of three
aortic valve implantation retrograde from the femoral artery. device generations of the corevalve revalving system. Circ Car-
Circulation 2006; 113:842–850. diovasc Interv 2008; 1:167–175.
36. Grube E, Laborde JC, Gerckens U, et al. Percutaneous implanta- 53. Roberts WC, Ko JM, Hamilton C. Comparison of valve structure,
tion of the CoreValve self-expanding valve prosthesis in high-risk valve weight, and severity of the valve obstruction in 1849 patients
patients with aortic valve disease: the Siegburg first-in-man study. having isolated aortic valve replacement for aortic valve stenosis
Circulation 2006; 114:1616–1624. (with or without associated aortic regurgitation) studied at 3
37. Bojara W, Mumme A, Gerckens U, et al. Implantation of the different medical centers in 2 different time periods. Circulation
CoreValve self-expanding valve prosthesis via a subclavian artery 2005; 112:3919–3929.
approach: a case report. Clin Res Cardiol 2009; 98:201–204. 54. Lefevre T. SOURCE Trial. In: TCT 2008. Washington, DC, 2008.
[29/6/010/15:46:49] [554–567]
43
Percutaneous therapies for mitral valve disease

Ryan D. Christofferson, Patrick L. Whitlow, E. Murat Tuzcu, and Samir R. Kapadia
INTRODUCTION Echocardiography is the cornerstone for diagnosis.

The prevalence of mitral valve disease, especially mitral regur- Transthoracic echocardiography (TTE) is most useful to
gitation (MR), is increasing. Despite significant gains in the image the subvalvular apparatus. Transesophageal echocar-
eradication of rheumatic fever, rheumatic mitral stenosis (MS) diography (TEE), especially with 3-D reconstruction, is very
remains a significant problem in underdeveloped countries. In helpful to identify commissural fusion. However, in good pro-
addition to the increased incidence of degenerative mitral valve portion of patients TTE provides adequate information in this
disease due to the aging population, the growing congestive respect. If the severity of MR is unclear, TEE is very helpful.
heart failure epidemic has led to further increase in the pro- The mitral valve splitability score (1) is the most common
portion of patients with severe MR. The emergence of new method of assessment for suitability for BMV, although there
percutaneous technologies for mitral valve disease has gener- are other validated methods (2,3).
ated great interest in the interventional community. Although
percutaneous balloon mitral valvuloplasty (BMV) is a well- Indications
established technique, the skill set necessary for this procedure Symptomatic Mitral Stenosis: ACC/AHA Guidelines
remains critical to the well-rounded structural heart disease According to the American College of Cardiology and American
interventionalist. This chapter reviews the indications and Heart Association (ACC/AHA) guidelines, class I indication for
techniques for percutaneous therapies for mitral valve disease. mitral valvuloplasty is the symptomatic patient with MS with
favorable anatomy for BMV. Asymptomatic patients with pul-
monary hypertension have class IIa and atrial fibrillation with
PERCUTANEOUS TREATMENT OF MITRAL
favorable anatomy has Class IIb indication. It is not appropriate
STENOSIS
to perform BMV in patients with nonrheumatic mitral valve
Fundamentals and Anatomic Considerations
disease and patients with prosthetic mitral valve stenosis.
Etiology
The most common cause of MS is rheumatic heart disease. Less
commonly, congenital MS may be detected in children. Rarely, Surgical Mitral Valve Replacement Vs. Percutaneous
MS is a result of collagen vascular disease, mucopolysacchar- Balloon Mitral Valvuloplasty
idoses, amyloid deposits, or is drug-induced. In rheumatic Surgical options for MS include close commissurotomy, open
mitral valve disease, fusion of the mitral valve apparatus may commissurotomy and mitral valve replacement. Close commis-
occur in the commissures, the cusps, or the chordae tendinae. surotomy skills are not available in the United States because
While the majority of patients have a combination of the above, this procedure is rarely indicated or performed (4). In devel-
30% of patients have isolated commissural thickening, 15% oping countries where BMV may be more expensive than
have thickening of the cusps only, and 10% have only chordal surgery, close commissurotomy can be considered for eco-
thickening. Generally, the mitral cusps thicken at the edges and nomic reasons if surgical expertise is available. Open commis-
fuse at the commissures, while the chordae thicken, shorten, surotomy is rarely indicated because if the valve is suitable for
and fuse. This leads to a funnel-shaped valve with reduced this procedure, it is suitable for BMV. The outcomes of BMV are
leaflet mobility and a fish-mouth shaped orifice (Fig. 43.1). If very comparable to surgical procedures even in randomized
the commissures are predominantly involved, this leads mainly trials and with long follow-ups (Fig. 43.2) (5). Valve replace-
to MS. Isolated thickening and shortening of the chordae results ment is reserved for patients who need mitral valve interven-
mainly in MR; however, if the cusps are thickened and adher- tion but are not candidates for BMV or BMV has failed.
ent, so they cannot adequately open or close, a combination of
MS and regurgitation occurs. Procedure
In the current era, BMV is exclusively performed using Inoue
Pathophysiology and Diagnosis Balloon (Toray International America, Inc., Houston, Texas,
Clinical manifestations of MS are caused by increased left atrial U.S.) in the United States and most of the world (Fig. 43.3).
pressure (shortness of breath), atrial fibrillation (thromboemb- Although BMV can be performed with double balloons, this
olism), or increased pulmonary pressures (fatigue). The pres- method is rarely used (6,7). A special metal dilator, the Cribier’s
ence of symptoms is the most important indication for dilator, has been used to reduce costs but because of technical
intervention. Gradients or size of mitral valve should not be challenges and a potential safety issue, it did not gain wide-
used to time intervention in asymptomatic patients with nor- spread acceptance (8).
mal pulmonary artery (PA) pressures. Exercise echocardiogra- Patient selection is the most critical step in the success of
phy at times is very useful to document functional status and the procedure. Patients with asymmetric commissural calci-
hemodynamic changes with exercise. fication, severe subvalvular scarring, significant MR, and
[29/6/010/15:46:49] [554–567]
PERCUTANEOUS THERAPIES FOR MITRAL VALVE DISEASE 555
Figure 43.1 A surgeon’s view of mitral stenosis. Intraoperative photograph of transseptal puncture site on the left, and surgical view of the
mitral valve post balloon valvuloplasty. The mitral valve has one fused commissure (arrow) and one open commissure (double arrows).
Figure 43.2 Mitral valve area following mitral commisurotomy by

BMC (BMC, n ¼ 30), OMC (OMC, n ¼ 30), or CMC (CMC, n ¼ 30).
Abbreviations: BMC, balloon mitral commisurotomy; OMC, open
mitral commisurotomy; CMC, closed mitral commisurotomy. Figure 43.3 A flouroscopic image of the Inoue balloon technique
Source: From Ref. 5. for mitral valvuloplasty (producer, origin). Note the simultaneous use
of transesophageal echocardiography to guide balloon inflation.
significant tricuspid regurgitation are not ideal candidates for of a pigtail catheter into the noncoronary sinus. The pigtail
BMV. TEE is very helpful at the time of the procedure to make catheter is useful in this position as a landmark to guide the
sure there is no left atrial clot and to assess the mechanism as transseptal puncture. If necessary for anatomic assessment
well as severity of MR after each balloon inflation. Online prior to transseptal puncture, the left groin venous sheath can
assessment of MR and commissures with 3-D TEE helps to be used to introduce an NIH or pigtail catheter into the right
optimize BMV, and hence helps to improve procedural out- atrium at the junction with the superior vena cava (SV). A
comes. Further, TEE can also help in making transseptal punc- power injection at this location will opacify the right atrium
ture safer. and pulmonary vasculature during the dextro phase, followed
It is standard procedure to perform BMV under the by the pulmonary veins and left atrium (LA) during the levo
guidance of invasive hemodynamic monitoring, requiring two phase. A PA catheter is always placed for PA pressure mon-
pressure transducers for simultaneous measurement of left itoring and assessment of cardiac output. The wedge position
atrial and left ventricular pressure. In addition, it is useful to can be used to evaluate left atrial pressure and monitor for
monitor PA pressure and cardiac output during the procedure. ventricular wave (V wave).
Therefore, it is necessary to have venous access in two locations An 8-Fr Mullins (Medtronic, Inc., Minneapolis, Minnesota,
and arterial access. The most straightforward configuration to U.S.) sheath is advanced to the SVC over the wire and then the
begin the procedure includes an 8-Fr short venous sheath in the wire is exchanged for the Brockenbrough needle (Medtronic,
left groin, a 4-Fr arterial sheath in the left groin, and an 8-Fr Inc., Minneapolis, Minnesota, U.S.). The sheath, dilator, and
short venous sheath in the right groin. needle are slowly brought down with some clockwise rotation.
The left groin arterial sheath is used to perform diagnos- Three drops are typically felt when the system crosses from SVC
tic cardiac catheterization, if necessary, and to allow placement to right atrium, limbus, and aorta. The location is confirmed by
[29/6/010/15:46:49] [554–567]
anteroposterior (AP) and lateral views to see the needle on the monitoring with TEE, fluoroscopy, and hemodynamics during
right border of spine in AP view and facing posteriorly away the procedure makes this technique safe and effective.
from the aorta in the lateral view just below the level of non-
coronary cusp. If there is good tactile pulsation of the LA, the
needle is advanced across the septum, with the hemodynamic PERCUTANEOUS TREATMENT OF MITRAL
tracing showing a change from a right atrial waveform to a left REGURGITATION
atrial waveform without any loss of pressure. If there is a sudden Introduction
drop in pressure during transit, this may indicate the needle has Significant morbidity and mortality can be attributed to MR. This
passed through the pericardial space, and should be withdrawn is true for patients with degenerative valve disease, as well as the
and pericardium should be carefully investigated. If the passage growing population of patients with functional MR. Although
is clean, the Mullins sheath can be advanced into the LA, and the surgical repair of MR has been shown to be superior to valve
dilator removed. Once the appropriate position of the sheath is replacement, a significant number of patients currently still
confirmed, heparin (70 U/kg) may be administered. Following undergo valve replacement. Percutaneous repair of the mitral
introduction of a coiled floppy Torray wire, the Mullins sheath is valve is under investigation with the potential to provide
removed and the septum is dilated for one minute with a long decreased morbidity, improved recovery time, and shorter hos-
11-Fr dilator. pital stays compared with open-heart surgery. These percuta-
Inoue balloon is sized according to height of the patient. neous techniques are predominantly based on existing surgical
A 26-mm balloon is used for patients up to 170 cm, 28-mm strategies, and each technique provides a different advantage
balloon for patients 170 to 180 cm, and 30-mm balloon for based on the anatomical and functional characteristics of the MR.
patients taller than 180 cm. Gradual dilatation with 1- to 2-mm Proper patient selection will ultimately determine the success of
increment should be used. Careful assessment of hemodynam- these emerging technologies. In addition, imaging technologies
ics (LA pressure, gradient, V wave, PA pressure), and TEE in and outside of the catheterization laboratory and integration of
assessment of valve (MR and MS) help one to decide how far to multiple imaging modalities will be important for safety and
push. Commissural MR indicates that commissural splitting efficacy of these percutaneous technologies. Finally, evaluation
has been achieved. Sometimes ventriculogram can help if the of MR devices will be challenging as they will be compared with
severity of MR remains questionable after TEE interrogation. traditional surgical techniques, which may have different goals
in patient management (e.g., need for repeat procedures, resid-
ual MR). These devices may possibly have a complimentary role
Clinical Aspects to surgery in the future.
The most important predictor of long-term success of BMV is the
procedural result. If MVA after the procedure is >1.6 cm2, the
long-term outcome is excellent (9–11). Some predictors of poor Fundamentals and Anatomic Considerations
outcome include age, atrial fibrillation, and valvular anatomy for Percutaneous Repair
(10–15). If the BMV is successful, PA and LA pressure will Mitral Regurgitation, Etiology, Diagnosis
decrease immediately and cardiac output will improve. Patients Normal mitral valve closure depends on the appropriate anat-
also feel an immediate difference in their exercise tolerance. omy and function of each component of the mitral valve,
Adequate control of heart rate and diuretic therapy are the corner including the annulus, the anterior and posterior leaflets, the
stone of medical management. It is also worthwhile to try and chordate tendineae and papillary muscles (17). Primary mitral
maintain sinus rhythm. Antiarrhythmic medications like flecai- valve disease refers to myxomatous degeneration of the leaflets
nide or sotalol can be very useful in these patients. Anticoagula- and chordae, resulting in chordal elongation or rupture, flail
tion is imperative in patients with a history of thromboembolism leaflet, or mitral valve prolapse. This is known as degenerative
or atrial fibrillation. Patients in sinus rhythm who have a large LA MR. Secondary valve disease (functional MR) is caused by
and are on b blockers (which can predispose them to asympto- disruption of the structural arrangement of the mitral appara-
matic atrial fibrillation) should be strongly considered for oral tus, most commonly resulting from dilated cardiomyopathy,
anticoagulation. Follow-up with symptom review and echocar- leading to ventricular dilatation, mitral annular dilation, and
diogram is recommended. Stress echocardiogram is helpful if altered papillary muscle-leaflet interaction. Moreover, ischemic
there is any question about worsening symptoms. cardiac disease may be associated with papillary muscle dys-
function or rupture, sometimes referred to as ischemic MR. In
this condition, changes in left ventricular geometry lead to
Limitations leaflet “tenting” and anterior leaflet override, which prevents
Patients with mixed MS and MR need mitral valve replacement proper coaptation (Fig. 43.4) (18,19).
and cannot be treated with BMV. Severe MR following BMV is In addition to the above categorization based on etiology
typically from flail mitral leaflet and usually requires surgery. and mechanism, a morphologic classification proposed by
Emergent surgery is rare but should be available because BMV Carpentier (Fig. 43.5) (20) groups the mechanism of regurgita-
is usually performed in very young patients with otherwise tion according to leaflet pathophysiology. This simplification
excellent prognosis even with emergent surgery. Restenosis has utility in terms of surgical and percutaneous approach, as
after BMV is 10% to 30% at five years depending on the initial the goal of therapy may be to restore normal leaflet function,
results (16). A repeat procedure can be attempted depending on but not necessarily normal valve anatomy.
the cause of restenosis or failure of the first procedure. Various methods exist for quantification of MR but echo-
cardiography is considered the gold standard. Direct measures
Conclusions including regurgitant orifice area, regurgitant flow, or diameter
BMV is the treatment of choice for patients with severe symp- of vena contracta constitute the major quantitative parameters
tomatic MS with favorable anatomy for the procedure. Proce- used in clinical trials. However, MR severity is judged after
dural success determines long-term outcomes. Careful taking into account many other secondary effects like size and
[29/6/010/15:46:49] [554–567]
Figure 43.4 Local left ventricular remodeling leading to mitral regurgitation by valve leaflet tethering, also known as functional or ischemic
mitral regurgitation. Abbreviations: LA, left atrium; LV, left ventricle; PM, papillary muscle; Ao, aorta.
Figure 43.5 The Carpentier classification of MR based on valve leaflet morphology. MR despite normal leaflet motion (type I, normal) results
from leaflet perforation. MR from myxomatous degeneration results from excessive leaflet motion (type II, excessive) with or without cordal
rupture and flail segment. Restrictive leaflet motion (type IIIa, restrictive A) is the most common cause of mitral regurgitation in rheumatic
heart disease with subvalvular apparatus scarring. Restrictive motion of mitral valve after myocardial infarction (type IIIb, restrictive B) from
tethering of leaflets causes MR with normal subvalvular apparatus. Abbreviation: MR, mitral regurgitation.
function of LA and left ventricle (LV), blunting of the pulmonary pulmonary hypertension in an asymptomatic patient can also
venous flow, serial changes in all these parameters over time, be an indication for surgery. Among patients with primary
and loading conditions. Further, postprocedural MR is even valve disease and severe LV dilation or dysfunction, the role
harder to quantify at times because of anatomical distortion of of surgery is controversial. In case of valve dysfunction second-
the valve from various procedures which may lead to very ary to LV dysfunction (functional MR), repair may be consid-
eccentric jets (e.g., after edge-to-edge repair). ered after medical optimization and cardiac resynchronization,
if indicated. There are several caveats to these general guidelines
General Indications for Invasive Treatment of Mitral and individualization of therapy is critical after considering
Regurgitation functional status, anatomy of the mitral apparatus, LV function,
In accord with guidelines released by the ACC/AHA, any and risk of surgery.
patient with symptomatic chronic severe MR should undergo Although no guidelines exist for determining the optimal
valve surgery (21). When the patient is asymptomatic, the utilization of percutaneous techniques, the surgical guidelines
development of LV dysfunction (ejection fraction <60%) or can serve as a template for the patient populations thought to
dilation (end-systolic dimension 40 mm) indicates the need be most likely to benefit from percutaneous repair. The indica-
for surgical intervention. In surgical centers that are experienced tions for percutaneous techniques, however, will likely differ
in mitral repair, surgery can be considered in asymptomatic from surgical indications given the nature of each approach
patients with less dilated LV and preserved systolic function, if and the possibility for early, less invasive intervention before
repair is feasible. The development of atrial fibrillation or development of sequelae such as LV dysfunction.
[29/6/010/15:46:49] [554–567]
Surgical Considerations of the proximity of the coronary sinus (CS) to the mitral
Reviewing the surgical approaches to mitral valve repair is annulus to create favorable geometric changes in the annulus,
useful because the majority of percutaneous repair techniques moving the posterior leaflet closer to the anterior leaflet,
attempt to mimic established surgical techniques. Therefore, an improving coaptation. This approach is termed indirect
understanding of the surgical approaches with their respective annuloplasty or CS reshaping. Additionally, there are devices
results and limitations is necessary to understand and evaluate under investigation that perform a direct annular repair by
emerging percutaneous techniques. plication from the left ventricular chamber. Other devices take a
The most common methods of surgical repair are based more direct approach to the annulus by applying suture-based
on restricting the mobility of the posterior leaflet (and at times plication systems. Cardiac chamber reshaping has been
prolapsing anterior leaflet) and reducing the size of annulus to explored as a means to bring the leaflets into coaptation by
achieve proper coaptation of leaflets. Therefore, these techni- decreasing the septal-to-lateral (SL) dimension of the ventricle
ques address the leaflets and annulus at the same time. Annu- or atrium. Finally, an early effort is underway to develop
loplasty is typically performed using a ring, although the choice percutaneous valve replacement. The devices selected for
of rings depending on the shape, stiffness, and extent of annu- review are those with some clinical or early clinical data,
lar coverage remains controversial (22–24). The classic leaflet although many other devices and approaches in development
repair techniques include triangular resection, quadrangular are not mentioned in this chapter.
resection, sliding annuloplasty, or edge-to-edge repair.
Although techniques of annuloplasty and leaflet repair are Clinical and Anatomic (Imaging) Screening for Each Class of Device
usually combined when treating degenerative valve disease, The dominant imaging modality for assessment of MR is echo-
isolated ring annuloplasty without leaflet repair is the domi- cardiography. Two-dimensional echocardiography is useful to
nant strategy for functional and ischemic MR (25,26). evaluate valvular structure, the presence of calcification, leaflet
Annuloplasty for functional MR results in improvement tethering, flail leaflet, or wall motion abnormalities. The severity
in NYHA class, favorable LV remodeling, and decreased of the regurgitation can be accurately quantified, as well as the
admissions for heart failure, but has failed to demonstrate a specific hemodynamic consequences of the regurgitation. In
mortality benefit (27). This may be in part due to the high most cases, a combination of clinical information and echocar-
recurrence rate of MR after isolated annuloplasty (28). The diographic data will establish the etiology of MR and the partic-
combination of annuloplasty and leaflet repair (Carpentier’s ular anatomy of the mitral valve, which is important in assessing
technique) for degenerative MR has shown the most favorable candidacy for surgical or percutaneous intervention. TEE can be
results in terms of mortality and recurrent MR. In experienced used to supplement the evaluation, often allowing greater def-
centers, Carpentier’s technique has provided results that are inition of mitral valve structures and regurgitation severity. This
better than with mitral valve replacement (29) and has resulted can be particularly useful to determine if edge-to-edge valve
in a mortality rate similar to that of the general population repair is feasible, as valve anatomy such as coaptation depth and
(86–93% survival at five years) (29–32). The recurrence rate of annulus size can be critical to clip feasibility.
severe MR (grade 3þ or 4þ) after repair is 3.7% per year (32). Other imaging modalities including cardiac computed
tomography (CT) (33), cardiac magnetic resonance (34), and
Classes of Percutaneous Mitral Valve Repair Devices 3-D echocardiography (35) may be useful to determine which
As a result of the variable anatomy and etiology of MR, a patient and which device is appropriate. In particular, indirect
number of different approaches have been developed for annuloplasty via the CS relies on proximity of the CS to the
percutaneous repair of magnetic resonance (Table 43.1). The valve annulus. Cardiac CT is particularly well suited to assess
goal of each approach is to bring the leaflets together to for this anatomical relationship, as well as proximity to the left
improve coaptation. The most advanced approach to date is circumflex artery, where such devices have the potential to
edge-to-edge leaflet repair, based loosely on the surgical repair cause ischemia. With cardiac MR it is possible to obtain signif-
championed by Dr. Alfieri. Another approach takes advantage icant structural information regarding the geometry of the LV,
mitral annulus and leaflets, and quantitative regurgitant vol-
umes (34). Finally, 3-D echocardiography may become impor-
Table 43.1 Percutaneous Mitral Valve Repair Technologies
tant in assessing valve characteristics and geometry, in addition
Edge-to-edge leaflet repair to the possibility of providing “real-time” guidance of percuta-
Evalve MitraClipTM (phase II) neous valve interventions, an application limited by current
Indirect (coronary sinus) annuloplasty technologic capabilities.
Viacor Percutaneous Transvenous Mitral Annuloplasty (phase I)
Edwards MONARCTM (phase I)
Cardiac Dimensions CARILLONTM (phase I) Clinical Aspects
Septal-to-lateral diameter reduction Percutaneous valve repair technology shows promise for
Myocor CoapsysTM (phase II)
reducing the morbidity and mortality associated with tradi-
Ample Percutaneous Septal Sinus Shortening (PS3) (phase I)
Direct transventricular (retrograde)
tional invasive surgical repair. The success of these techniques
Mitralign Mitralign (phase I) is dependent on proper selection of patients according to the
Guided Delivery Systems Accucinch (preclinical) clinical presentation and the anatomy along with appropriate
Valve/implant modification imaging of the valve before and at the time of procedure. The
QuantumCor RF annuloplasty (preclinical) major advantage of a percutaneous approach, beyond its less
MiCardia Dynamic Annuloplasty Ring (preclinical) invasive nature, is the ability to perform procedures in a beat-
Cardiosolution Percu-Pro System (preclincal) ing heart, which allows immediate evaluation of the results. If
Valve implant percutaneous technologies can be made simple, relatively fail-
Endovalve (preclinical) safe, and preserve surgical options, they represent the future of
[29/6/010/15:46:49] [554–567]
valve intervention. Proper training of the interventionalists not feasibility trial has been completed, with short-term and
only in device manipulation but also in patient selection, six-month results in the first 27 patients published in 2005
imaging techniques, and catheter skills will be necessary for (39). This study enrolled patients with moderate to severe
wide adaptation of these techniques with favorable results. MR (3þ or 4þ by core laboratory evaluation) who were
candidates for traditional open surgery in the event of com-
Edge-to-Edge (Double-Orifice) Leaflet Repair plications. Exclusions included patients with rheumatic or
An isolated edge-to-edge repair, championed by the Italian infectious MR, recent surgical or interventional procedure,
surgeon, Alfieri, has been shown in a small series to have dilated ventricle, severe ventricular dysfunction, severe annu-
reasonable long-term results (36). This repair technique was lar calcification, or mitral orifice area <4 cm2. On echocardio-
the basis for one of the earliest applications of surgical repair gram, the MR needed to occur between A2 and P2 and meet
techniques to percutaneous valve intervention. Like its surgical certain parameters for flail dimensions or leaflet tethering so
counterpart, the percutaneous edge-to-edge procedure produ- that the device could be expected to capture the leaflets
ces a double-orifice and a fibrosing bridge segment without the adequately. Degenerative valve disease was the leading etiol-
development of significant MS (37). Early in the inception and ogy (93%), although there were some ischemic patients (7%).
development of this technique there were two major devices Successful clip placement was achieved in 24 (89%)
with significant preclinical and clinical data. However, the patients. Three patients had clip placement but not enough
device developed by Edwards Lifesciences was a suture- reduction in MR, and went on to successful surgical repair.
based device, and development has been halted because of The complication rate was low, and all patients were discharged
technical difficulties. The device with continued clinical activity home after an average of 2.5 days. In follow-up there was partial
is the MitraClipTM Endovascular Cardiovascular Valve Repair clip detachment in three patients, necessitating elective surgical
System (CVRS) developed by Evalve, Inc. (Menlo Park, Cali- repair. An additional six patients had moderate or severe MR,
fornia, U.S.). This v-shaped clip device is delivered through a with two patients undergoing elective repair or replacement. Of
24-Fr steerable guide catheter by transseptal approach the 27 original patients, 13 (48%) received a clip successfully and
(Fig. 43.6) (38). The clip is used to grasp the leaflets from remained with MR severity 2þ at six months of follow-up. The
beneath the valve, creating the double-orifice repair. Position- protocol was modified midway through the study allowing two
ing is confirmed by echo and fluoroscopy and the clip is locked clips to be placed, which may have improved the subsequent
into position. If needed, the clip can be reopened and reposi- success rate for MR reduction. With respect to safety, only 15% of
tioned prior to release. patients experienced a major adverse event, including the three
Initial data from a porcine model was published in 2003 clip detachments and one permanent stroke. This was signifi-
showed that in adult pigs with no MR (n ¼ 14), a functional cantly lower than the 34% predicted for a similar surgical cohort.
double-orifice could be successfully created by endovascular The EVEREST I feasibility registry completed enrollment
placement of a mitral valve clip (38). In the vast majority of the with a total of 55 patients. A pivotal phase II trial (EVEREST II)
pigs, clip attachment was durable with no clip embolization. has also completed enrollment, and compares the endovascular
Progressive healing was observed so that a mature, continuous CVRS approach with standard cardiac surgery. The study design
bridge of tissue developed between the anterior and posterior is a prospective, multicenter, randomized controlled trial with a
leaflets at the site of the clip with complete endothelialization 2:1 randomization to study and control arms, respectively. The
and encapsulation of the clip. study has two primary endpoints, safety and efficacy. The pri-
Endovascular Valve Edge-to-Edge Repair Study mary efficacy endpoint is noninferiority to cardiac surgery with
(EVEREST) I, a phase I prospective, multicenter safety and respect to a composite endpoint of freedom from surgery for
valve dysfunction, death, and MR 2þ. The safety endpoint is
superiority of an endovascular strategy over surgical valve repair
or replacement, defined as freedom from 30-day major adverse
events. This is a unique study because it is the first randomized
trial comparing surgical therapy with percutaneous device
implantation for valvular heart disease with an independent
core laboratory evaluating MR. A total of 279 patients have
been enrolled in this study from 35 centers in the United States
and Canada.
Follow-up data has been made available at major national
meetings from the EVEREST I feasibility study and non-
randomized, roll-in EVEREST II patients. The most recent
reporting on 107 patients deemed the “EVEREST preliminary
cohort” (55 patients from EVEREST I and 52 patients from
EVEREST II) shows a success rate of 76% for clip implantation.
If acute procedural success was obtained, results were durable
with 63% of patients remaining free from death, surgery, or
MR > 2þ at 36 months of follow-up. Death in follow-up was
Figure 43.6 The Cardiovascular Valve Repair System with the very uncommon (4%), and most patients (82%) remained free
MitraClipTM, produced by Evalve, Inc. (origin). Clip is inserted via the from surgery at 36 months. Significant LV reverse remodeling
transseptal approach (large panel) with inset showing magnified has been demonstrated when the clip successfully reduces the
view of the clip (top right) and proximal handle of the device MR grade to 2þ. The 30-day major adverse event rate was
(bottom). low at 9%, with only one patient requiring prolonged ventila-
tion (>48 hours), one stroke, and one death unrelated to the
[29/6/010/15:46:49] [554–567]
clip procedure. In the subgroup of patients with functional patients from the EVEREST I and roll-in patients from
MR (n ¼ 23), similar results have been noted with 87% freedom EVEREST II, showing 67% were successfully repaired and
from major adverse events at 30 days, 80% freedom from severe 33% replaced. Finally, it remains to be seen if this device is
heart failure (NYHA class 3), and 79% freedom from severe useful only for degenerative disease, or whether it can also be
MR (>2þ) at one year. used for functional MR.
Preliminary results from the high-risk registry have also
been reported at a major national meeting but are as yet Indirect Annuloplasty via Coronary Sinus
unpublished. The registry was designed to demonstrate the The proximity of the CS to the mitral annulus has provided the
superiority in terms of safety of the clip to traditional surgery in opportunity for a creative approach to reduction of MR. The
high-risk patients by comparing the cohort with the predicted indirect annuloplasty approach utilizes devices placed in the
morbidity and mortality from the Society of Thoracic Surgeons CS to modulate the shape and size of mitral annulus. This
(STS) database risk predictor model. The patients enrolled (n ¼ approach is promising given the fact that most patients under-
78), unlike the EVEREST I and II cohorts, were mostly func- going surgical repair have an annuloplasty either alone (as in
tional (64%) and fewer degenerative MR patients. The success functional MR) or in conjunction with other repairs (degener-
rate for clip implantation was high (96%) and two clips were ative MR). The challenges to this procedure include anatomic
placed in over one-third of patients. The primary safety end- variability in the location of the CS with respect to the mitral
point was met, with 7.7% 30-day mortality in the clip cohort, annulus (42) and the proximity of the left circumflex coronary
with a predicted mortality of 18.2% (p < 0.008). Of those artery (Fig. 43.7) (42,43). Imaging techniques, such as CT
surviving, the NYHA class improved in 76%, with three- angiography and echocardiography may be useful to identify
fourths of patients having NYHA Class I-II symptoms at suitable candidates for these procedures. The long-term success
30-day follow-up. of this approach will also depend on the long-term safety of CS
In addition to the determination of safety and efficacy as instrumentation (e.g., displacement of device or forces, throm-
defined in the pivotal trial, additional questions need to be bosis, perforation) and the need for other devices in the CS (e.g.,
answered with the Evalve CVRS prior to widespread adoption biventricular pacing).
of this procedure. One important issue is the prolonged proce- Viacor Percutaneous Transvenous Mitral Annuloplasty.
dure time (204 116 minutes) and steep learning curve for The Percutaneous Transvenous Mitral Annuloplasty (PTMA)
device implantation. It appears from unpublished EVEREST I system was developed by Viacor, Inc. (Wilmington, Massachu-
data that compared with first procedures (device time for one setts, U.S.) (Fig. 43.8). This device utilized implantable rods to
clip ¼ 181 minutes), subsequent procedures for an individual reshape the CS and mitral annulus by impinging on the
operator have a much shorter duration (134 minutes) (40). annulus and decreasing the septal-lateral mitral annular diam-
Similar experience has been noted with approval of this device eter. This change in annular geometry increases leaflet coapta-
in Europe where careful training has resulted in reasonable tion to reduce or eliminate MR. The device is composed of
procedural time and good success. A second issue is whether nitinol and stainless steel rods coated with Teflon and plastic,
the clip will eliminate subsequent surgical options, thus failing and ranges from 35 to 85 mm in length. It has a rigid distal
the “nothing to lose” standard. This question was addressed in element, connected to a flexible push rod for delivery. Via
a publication detailing the initial experience in the first six subclavian venous access, the device is passed through the
patients undergoing surgery after clip placement, showing that lumen of a guiding catheter, causing a conformational change
in all cases the clip was uneventfully removed and the surgical in the annulus that can be assessed in real time by echocardiog-
options were not limited (five repairs and one replacement) raphy. Rods of varying length and stiffness are implanted in
(41). Similar unpublished data are now available for 36/107 sequence up to a maximum of three rods, until the optimal
Figure 43.7 Computed tomography images of the anatomic relationships near the mitral annulus. The left panel shows the relation of MA,
CS, and the circumflex artery. The LCX crossed between the CS and MA in 80% of patients studied, with significant variability in location of
crossing (CS/LCX intersection). 2C, 3C, and 4C refer to the typical two-, three-, and four-chamber views used in echocardiography. The right
panel shows that the distance between the CS and MA is less anteriorly and posteriorly but maximum in the lateral aspect (arrow).
Abbreviations: MA, mitral annulus; CS, coronary sinus; LCX, left circumflex artery. Source: From Ref. 42.
[29/6/010/15:46:49] [554–567]
decreased from 6.5 2.2 to 0.4 0.5 cm2 (p < 0.03), and the
vena contracta width was reduced from 8.2 4.7 to 2.6 0.9 cm2
(p ¼ 0.03). The mitral annular diameter went from 30 2.1 to
24 1.7 mm (p ¼ 0.03) following device placement. There was no
MS induced by the device, and left ventricular ejection fraction
was improved.
A subsequent sheep experiment using 3-D echocardiog-
raphy assessed the effect of percutaneous placement of up to
three annuloplasty devices of varying size and stiffness in the
custom multilumen CS catheter (45). At eight weeks following
experimental induced posterior myocardial infarction, the
annuloplasty device significantly reduced the MR jet area, the
mitral annular A-P dimension in systole and diastole, and
mitral valve tenting area in all three planes.
Human implantation began initially on 10 patients under-
going open heart surgery for functional MR (46). The study
included the single lumen prototype device and the multilumen
next-generation device. Successful implantation and removal
took place in nine patients, confirming the feasibility of alteration
of annular geometry and MR. Subsequently the Percutaneous
Transvenous Mitral Annuloplasty (PTOLEMY) I study has been
completed, the initial human feasibility trial with permanent
implantation. The results of this trial are not yet published, but
have been presented orally at national meetings. The study
included 31 patients, and showed successful implantation in
11 patients. A single device fracture, one CS perforation, two
CS dissections, and two device slippages occurred. The MR
reduction was on average one grade (MR 3þ to 2þ), and 3-D
TEE showed significant annular modification in patients with
successful implant.
The PTOLEMY I experience was valuable to improve the
device, with changes in the rod to improve push (reduction in
AP diameter of the mitral annulus) and changes in the suture to
decrease slippage. Device design is now frozen, and a second
feasibility and efficacy trial will begin (PTOLEMY II). This trial
will enroll 60 patients in the United States and abroad and will
Figure 43.8 The Viacor Percutaneous Transvenous Mitral Annu- have primary endpoints of six-month freedom from major
loplasty (PTMA) system (origin). The system features an external adverse events and MR reduction efficacy. In addition, second-
sheath depicted below with implantable rods that provide tension to
ary endpoints will include reverse remodeling, quality of life,
the mitral annulus, improving leaflet coaptation and mitral regurgi-
tation. The delivery sheath on the top and three implantable rods
and six-minute walk test. Patients must have moderate func-
with different strengths in the bottom (top). The proximal end of the tional MR, symptomatic heart failure, and left ventricular
device (middle). The rods can be inserted depending on the need dysfunction. Patients may be excluded on anatomic or clinical
and response of the device in displacing posterior leaflet anteriorly concerns, including left dominant coronary circulation,
(bottom). planned bypass surgery, or prior mitral repair.
Advantages of the Viacor PTMA device include the ease
of placement, the number of combinations of devices of varying
lengths and strengths that can be used to optimize the reduc-
configuration is determined. The guide catheter is removed and tion in MR, and the ability to return later and change the device
a permanent implant is placed and capped. The cap is located to further reduce MR if it recurs. The major drawback to the
in a subcutaneous pouch, like a pacemaker, and can be reac- device includes the limitations inherent to a CS device: CS
cessed at a later date for revision of implants as needed for MR proximity to the mitral annulus and left circumflex artery.
reduction. There have been no reports to date, however, of left circumflex
A sheep model of ischemic MR, induced by experimental artery ischemia with this device.
circumflex artery ischemia, was used for early device experi- CARILLONTM Mitral Contour System. Under devel-
mentation (44). Within one minute of circumflex occlusion, all opment by Cardiac Dimensions (Kirkland, Washington, U.S.),
animals developed 3þ to 4þ ischemic MR. A single annuloplasty the CARILLON Mitral Contour System is a fixed length, dou-
rod was placed, its impact on the MR assessed, then circumflex ble-anchor device (Fig. 43.9) that is advanced through a catheter
flow was reestablished and the device removed. This cycle was and positioned in the CS. After the device is deployed and
repeated up to five times in each animal, using devices of locked into position, tension between the proximal and distal
increasing length of the stiff distal segment to find the optimal anchors of the device results in tissue plication and reduces
device. The results were reported only for the optimal device mitral valve annular diameter, resulting in decreased MR. Via
placed in each animal. Device placement resulted in reduction of jugular venous access, the nitinol annuloplasty device is posi-
MR to 1þ to 2þ in all animals In addition, the MR jet area tioned in the CS after cannulation and measurement of the CS,
[29/6/010/15:46:49] [554–567]
tachycardia-induced cardiomyopathy model (49), showing sim-

ilar one-month results to the canine model, with respect to
mitral annular diameter percent reduction (23.7 1.45) and
percentage MR jet area to left atrial area (27.84 1.96 before
procedure, decreased to 2.40 0.98 post procedure, p < 0.05).
There were no premature deaths in the ovine study. However,
in the sheep the left circumflex artery does not run in the
atrioventricular groove, precluding an assessment of safety
with respect to ischemia.
Early human experience was obtained on the initial
device design, but some failures due to distal anchor slippage
led to a change in device design with a larger distal anchor and
interlocking wire forms to improve device stability (CARIL-
LON XL). On the basis of this experience, a multicenter human
safety and feasibility study has been completed in Europe
entitled AMADEUS. This trial enrolled 48 patients with 2þ to
4þ functional/ischemic MR and NYHA Class II-IV. The pri-
mary endpoint was 30-day safety with secondary efficacy
endpoints of MR reduction, functional class improvement,
Figure 43.9 The CARILLON Mitral Contour System, developed by
quality of life, and six-minute walk test. This trial has been
Cardiac Dimensions, Inc. (Kirkland, Washington, U.S.). The device
consists of the distal and proximal anchor with a connector that reported in preliminary form at a major national meeting. This
synchs the annulus (top). The device is inserted via the SVC in the report showed successful implantation in 30 patients (63%),
CS and positioned depending on the anatomy of the CS, circumflex with one death, three myocardial infarctions and three CS
artery and the effect on mitral regurgitation as tension is applied to perforations at 30-day follow-up. There were no device embo-
synch the mitral annulus (bottom). Abbreviation: CS, coronary sinus. lizations and no patients needed surgery. The total major
adverse event rate was 13%. The coronary arteries were crossed
by the device in 84% of cases, but only prohibited device
placement in 14% of cases. In the remainder of cases the device
avoiding contact with the left circumflex artery if possible. The did not cause coronary impingement or was able to be recap-
distal anchor of the device is deployed and tension is placed on tured and repositioned to avoid coronary impingement. MR
the delivery system, creating tissue plication. The amount of was reduced by 27%, and functional class was reduced at six-
tension is manipulated as necessary to allow optimal reduction month follow-up by about one class, from average 2.86 to 1.76
in annular dimension and MR, as verified by concomitant (p < 0.001). Six-minute walk tests improved by about 100 m
echocardiography. When optimized, the proximal anchor is (307–403 m, p < 0.001), and there appeared to be no correlation
deployed and locked into position. The device can be recap- between the distance of the annulus and patients experiencing
tured by advancing the delivery catheter over the device, acute reduction in MR.
collapsing the device anchors. The advantages of the CARILLON device include the
The earliest preclinical testing with this device was done in simplicity of design, the adjustable positioning, and ease of
a canine tachycardia-induced cardiomyopathy model, with both delivery. Disadvantages include the same anatomic disadvan-
acute and chronic (four-week) hemodynamic evaluation of the tages to any device using the CS, particularly the variability in
device (47). The early canine experience highlighted some of the the relation of the left circumflex artery and potential to
anatomical, design, and safety issues, as 3 of 12 dogs had coro- induce ischemia. However, it appears that lack of proximity
nary anatomy that precluded placement without left circumflex of the sinus may not correlate with device efficacy, and the
ischemia (including two fatalities), and 2 of 12 dogs could not get direction of the “pull” from device may be a more important
a device because of inadequate size of the early prototype. Of the determinant of success, although this observation needs further
seven dogs with a successful implant, at four weeks there was a studies.
reduction in mitral annular size compared with those MONARCTM Percutaneous Transvenous Mitral Annu-
with unsuccessful implant (3.37 0.23 vs. 3.73 0.11 cm, loplasty System. Edwards Lifesciences, Inc. (Irvine, California,
respectively), as well as percentage MR jet area to left atrial U.S.) originally developed this annuloplasty device as the
area (0.11 0.04 vs. 0.39 0.05, p < 0.05). VIKINGTM system, but has subsequently marketed the second
A subsequent ovine model of experimental tachycardia- iteration of the device as the MONARC system (Fig. 43.10). The
induced heart failure demonstrated the acute hemodynamic original device consisted of a distal self-expanding anchor, a
effects of the device (48). Each sheep (n ¼ 9) was paced in the spring-like “bridge” segment, and a proximal self-expanding
ventricle for five to eight weeks at 190 beats/min, following anchor. The bridge element showed a tendency to fracture and
which MR of moderate severity was confirmed by echocardiog- so was redesigned to prevent this occurrence. The device is
raphy. The device was successfully placed in all animals, with a placed by advancement of a 12-Fr delivery catheter and device
significant reduction in mitral annular diameter from 4.17 into the CS. Following confirmation of proper device location,
0.14 to 3.24 0.11 cm (p < 0.001) and percentage MR jet area to including coronary angiogram, the distal anchor is deployed in
left atrial area from 41.9 6.4 to 4.1 2.8 (p ¼ 0.003). MR was the great cardiac vein (GCV) and the proximal anchor is
essentially absent in seven of nine animals after device place- deployed in the proximal CS. The bridge segment is designed
ment. The cardiac output increased and pulmonary capillary with shape-memory properties that lead to shortening of the
wedge pressure and mean PA pressure were reduced. The device at body temperature. The device cannot be recaptured
chronic effect of the device was also studied in the ovine after the anchor has been deployed.
[29/6/010/15:46:49] [554–567]
with recent ischemia or planned intervention were excluded, as

were patients with an implanted cardiac defibrillator (ICD) or
pacing leads in the CS. Other exclusions included patients with a
low ejection fraction (<25%), mitral valve prolapse, or moderate
to severe mitral annular calcification. The primary safety objec-
tive of the study was procedural success and 30- and 90-day
safety. Acute procedural success was defined as device implan-
tation without occurrence of in-hospital death, tamponade, or
myocardial infarction, and was achieved in 82% of patients. The
30-day safety endpoint defined as freedom from death, tampo-
nade, or myocardial infarction occurred in two patients who had
tamponade, but no deaths or myocardial infarctions occurred. In
long-term follow-up, however, three patients had myocardial
infarctions, all thought to be device related, with one leading to
death. There were nine unrelated deaths, leading to an overall
70% event-free survival at two years. The 90-day efficacy end-
point of reduction in MR by one grade was met, with mean MR
grade going from 2.48 to 1.78 (p ¼ 0.0002). There were some
favorable changes in ventricular dimensions and hemodynamic
parameters. Overall NYHA functional class was reduced from
2.66 at baseline to 2.0 at follow-up. Responders did better if the
Figure 43.10 The Edwards MONARCTM CS annuloplasty device initial MR grade was 3þ to 4þ.
(Edwards Lifesciences, Irvine, California, U.S.). The device is The EVOLUTION study is the largest human cohort of CS
implanted in the CS (top). The bridging element has shape memory implants, and interim results from this study shed some light on
and provides tension to the posterior mitral annulus, bringing the potential device-related issues over time. The procedure is
leaflets into coaptation, reducing mitral regurgitation. Abbreviation: straightforward and reproducible, and early efficacy data sug-
CS, coronary sinus. gest that moderate to severe MR patients are the most likely to
benefit from this device. The preliminary experience in the mild
to moderate MR (grade 2þ) patients is intriguing given the
interest in the use of percutaneous devices for less severe MR
Initial results in humans were reported for five patients to prevent adverse remodeling of the LV. Although data regard-
with chronic ischemic MR (50). Implantation of the device was ing LV parameters are not yet available, results from the EVO-
successful in four patients, with one failure due to difficulty in LUTION study indicate that there is no MR reduction in these
advancing the device, leading to perforation of the anterior patients. It remains to be seen if there are long-term clinical
interventricular vein and pericardial effusion. A subsequent benefit to early implantation of the device for prevention of MR
reattempt was also unsuccessful. The patients were followed progression. The EVOLUTION II trial will soon begin enrollment
for 180 days with serial exam, chest x-ray, and echocardiogram. and should answer some or all of these questions.
There was one late death (day 148) due to progressive heart
failure unrelated to the device. Coronary angiogram at 90 days Septal-to-Lateral Annular Cinching (Transatrial or Transventricular)
showed no evidence of circumflex artery compromise and the Geometric alterations in the LV and mitral annulus are a result
CS remained patent in all three surviving patients. Separation of left ventricular enlargement from dilated or ischemic cardi-
of the device bridge element was documented on follow-up omyopathy. This leads to MR labeled as functional/ischemic
chest x-ray in three patients (days 22, 28, and 81). Although MR as detailed above. The observation that the SL or AP mitral
migration of the anchors was not observed and no adverse annular diameter is increased in a tachycardia-induced model
clinical events occurred in these patients, the feasibility study of functional MR (51) has led to the development of a new
enrollment was discontinued. There was no significant change surgical technique for prevention of MR. This technique,
in the mitral annulus diameter, NYHA failure class, or MR known as septal-to-lateral annular cinching (SLAC), is based
grade at follow-up. on reduction in the SL annular diameter. This method has the
Following this early experience, the device was rede- potential to address some limitations of typical ring annulo-
signed, and is now called the MONARC system. The nitinol plasty, namely the fact that the trigone, thought to be fixed in
bridging segment was replaced with a “delayed-release system,” size, is not actually constant—it also dilates with rest of the
which utilizes a slow conformational change in the bridge ele- annulus. Therefore, a ring that does not address this part of the
ment to shorten the distance between the proximal and distal annulus is only partially successful, especially in ischemic
anchors. The shape change occurs due the slow breakdown of a patients. The SLAC approach may overcome this problem
biodegradable polymer in the bridge segment. This shortening because it remodels the ventricular apparatus. Initially, there
induces a conformational change in the CS, extending to the were two major percutaneous approaches to this method, one
mitral annulus, further reducing any postprocedural MR. involving placement of a transventricular chord (iCoapsysTM)
The EVOLUTION trial is the first multicenter feasibility and a second involving placement of a chord from the intera-
and safety study for the MONARC system and has completed trial septum to the CS (PS3). The iCoapsys device trial was
enrollment in Europe and Canada. Preliminary results have been recently discontinued because of a lack of funding with closure
presented at major national meetings. This study enrolled of the Myocor company. Edwards Lifescience has acquired the
72 patients with functional/ischemic MR, grade 2þ to 4þ, with intellectual property from Myocor but has not indicated that
appropriate CS dimensions to fit device specifications. Patients they are going to pursue device development at this time.
[29/6/010/15:46:49] [554–567]
Figure 43.11 The PS3 system, developed by Ample

Medical, Inc. (Foster City, California, U.S.) pulls two
sides of the annulus by placing an anchor in the CS and
interatrial septum with a rod across the atrium. (A) The
elements of the PS3 system consist of a CS device
(“T-Bar”) and an interatrial SD linked by a connecting
bridge which spans the SL axis of the MV. By applying
traction on the bridge, the devices are brought toward one
another, reducing the SL dimension. (B) Postmortem
radiography of an implanted PS3 system demonstrating
the T-Bar and SD straddling the SL axis of the MV.
Abbreviations: MV, mitral valve; CS, coronary sinus;
SL, septal-lateral; PS3, Percutaneous Septal Sinus Short-
ening; SD, septal device. Source: From Ref. 52.
PS3 system. The Percutaneous Septal Sinus Shortening 25.3 0.8 mm after 30 days (p value not given). The results
(PS3) system, developed by Ample Medical, Inc. (Foster City, for reduction in MR in the short-term were similar, with an MR
California, U.S.) pulls two sides of the annulus by placing an grade of 2.1 0.6 before procedure versus 0.4 0.4 post
anchor in the CS and interatrial septum with a rod across the procedure (p < 0.001). This result was maintained at 30 days
atrium (Fig. 43.11). The anchors are composed of nitinol and (MR grade 0.2 0.1). Additional hemodynamic and laboratory
supplied by AGA Medical. data were consistent with improved cardiac function.
Using a 12-Fr sheath in the right internal jugular vein, a The first human implant was recently performed without
catheter with incorporated magnet on its tip called the GCV complication and the case was presented at TCT 2006 (54).
MagneCathTM (Ample Medical, Inc., Foster City, California, U. Acute human studies have been reported in patients outside
S.), is advanced into the CS and positioned behind the posterior the United States with significant (up to 35%) reduction in
mitral leaflet. The left atrial (LA) MagneCath is introduced annular dimensions but no long-term testing has been reported.
through the Mullins sheath placed in the LA via right femoral The advantages of the PS3 system include relative ease of
vein after transseptal puncture. The catheter from the LA is placement, avoidance of circumflex coronary artery impinge-
advanced until it magnetically mates with CS catheter. A ment, and potentially safer use of transseptal puncture when
crossing catheter, introduced via the LA MagneCath, is compared with the similar transventricular cinching device
advanced into the GCV MagneCath, making a small hole in (discussed below). However, the transventricular device may
the left atrial wall. A glide wire is then advanced from the LA have additional advantages in terms of LV remodeling.
MagneCath into the CS and remains as a continuous external- CoapsysTM and iCoapsysTM. Myocor, Inc. (Maple
ized loop across the LA after both MagneCaths are removed. A Grove, Minnesota, U.S.) developed a device termed Coapsys,
T-bar element (CS anchor) is placed over the wire into the CS, which was surgically implanted with two pads in the pericar-
and an attached suture is pulled across the LA and externalized dium (anterior and posterior) with a wire passing through off-
at the right common femoral vein. Advanced over the suture pump with a cord placed between the pads through the LV.
element, the septal anchor (an Amplatzer patent foramen ovale This cord is then cinched up to decrease the SL diameter and
occluder, Golden Valley, Minnesota, U.S.) is deployed in stan- eliminate MR. In initial animal studies using a canine tachy-
dard fashion. Tension is applied on the suture to cause septal- cardia-model of functional MR (n ¼ 10), this device reduced the
lateral shortening, improving leaflet coaptation and reducing mean MR grade from 2.9 0.7 to 0.6 0.7 (p < 0.001), without
MR. The suture is locked into place with tension element, adverse consequence on ventricular function (55,56). A
completing the procedure. The company is in the process of randomized trial (RESTORE-MV) in humans is enrolling
changing the septal anchor to more specific device for this patients with coronary artery disease and ischemic MR, com-
application. paring traditional open CABG and mitral repair with CABG
This device has been applied to an ovine model of and Coapsys device placement. Intraoperative results from this
tachycardia-induced cardiomyopathy created by rapid right trial have been reported in the first 19 patients receiving the
ventricular pacing (53). The degree of reduction in functional implant, showing a reduction in MR grade from 2.7 0.8 to
MR, and in the septal-lateral systolic distance, was the primary 0.4 0.7 after implantation (p < 0.0001) (57). All implants were
efficacy measure of this study. Sheep underwent short-term performed successfully without cardiopulmonary bypass and
(n ¼ 19) and long-term (n ¼ 4) evaluation. The device was no hemodynamic compromise or structural damage to the
successfully placed in all animals with no evidence of circum- mitral apparatus.
flex coronary artery impingement and maintaining patency of Myocor subsequently developed a system to deploy this
the CS. The short-term results indicated a significant reduction device totally percutaneously without thoracotomy and called
in septal-lateral diameter from 32.5 3.5 mm before procedure it iCoapsys (Fig. 43.12). A specifically designed needle, guide-
to 24.6 2.4 mm post procedure (p < 0.001). This was main- wire, and sheath were used to obtain controlled access in to the
tained at 30 days in the long-term animals where the septal- pericardial space. Steerable suction-based catheters with intra-
lateral diameter was 30.4 1.9 mm before procedure and cardiac echocardiographic guidance were positioned on the
[29/6/010/15:46:49] [554–567]
(Mitralign, Salem, New Hampshire, U.S.) uses a device com-

posed of three metal anchors, connected by standard suture
material. The anchors are placed in the mitral annulus and the
suture is cinched to perform the annuloplasty.
The device is placed via arterial access, using direct
ventricular access with a unique translation catheter of trident
design for device delivery. The initial device design used a
magnetic guiding catheter placed in the CS, however, the new
device design is guided by standard imaging techniques. After
proper positioning of the guiding catheter retrogradely in the
LV, radiofrequency energy is used to cross the annulus and
three anchoring wires are deployed. Each anchor is attached to
a surgical suture and is threaded to the locking device. The
sutures are then cinched, directly plicating the annulus, leading
to reduction in MR.
Advantages of this device include direct placement on the
mitral valve annulus, avoiding the concern for left circumflex
artery compromise. The device is adjustable to deliver the
appropriate amount of tension to the annulus, and because of
its design, there are no sizing issues, as a single device is
intended for use with all patients. The CS remains free of
permanent implant, allowing for subsequent placement of CS
Figure 43.12 Myocor, Inc. (Maple Grove, Minnesota, U.S.) devel- leads for biventricular pacing. Long-term durability will be an
oped a device termed iCoapsysTM, which was percutaneously issue to follow with this device. The Mitralign pilot study in
implanted with two pads in the pericardium (anterior and posterior) humans is currently enrolling patients in Germany and Canada.
with a wire passing through offpump with a cord placed between the GDS Accucinch Annuloplasty System. Another device
pads through the left ventricle. This cord is then cinched up to is under development, similar to the Mitralign device described
decrease the septal-lateral diameter and eliminate mitral regurgita- above, using a catheter-based transventricular approach to
tion. Abbreviations: MV, mitral valve; RV, right ventricle; LV, left place anchors in the myocardium directly beneath the mitral
ventricle; Aov, aortic valve. Source: From Ref. 58.
annulus, performing a plication annuloplasty. A startup com-
pany called Guided Delivery Systems (GDS) has begun pre-
clinical studies using the device, termed the GDS Accucinch
Annuloplasty System. Preliminary data on this device is
anterior and posterior ventricular wall. The posterior target expected to be available soon.
zone is between the papillary muscle and the P2 segment of the QuantumCor. The application of radio frequency to
mitral annulus, about 2 cm apical to the atrioventricular groove. remodel the mitral annulus is under development by Quan-
Once proper alignment is achieved using fluoroscopy, a needle tumCor (Lake Forest, California, U.S.). The concept is termed
is passed from each catheter into the ventricle. A flexible wire transventricular annulus remodeling, and relies on scarring
introduced through the posterior catheter is captured by a and shrinkage of the mitral annulus after application of radio
snare from the anterior catheter. The flexible wire is used to frequency energy directly to the annulus. The device is
place the transventricular cord, which is exteriorized through intended both for surgical and transcatheter use (transseptal)
the delivery sheath. Then permanent implant device is placed and has a malleable tip with eight electrodes to deliver energy.
over the cord, posterior pad first. The cord is tightened to The catheter is connected to a pulse generator that is modulated
achieve the desired effect and is left in place. This device was by temperature sensors in the electrodes to regulate the amount
used in two patients successfully. However, because of finan- and time of energy delivered. The catheter can be manipulated
cial reasons the company was closed with termination of fur- to deliver radio frequency energy to specific locations on the
ther development of this device. annulus, allowing for adjustment of the procedure to individ-
ual anatomy. The device has been tested in the chronic sheep
Transventricular (Retrograde) Direct Annuloplasty model (n ¼ ?) showing a 21.4% reduction in annular diameter at
Although there is no published data available on devices 30 days and further 26% reduction in diameter over 180 days. A
placed on the valve annulus by retrograde transventricular human study should start in 2010.
approach, these devices will soon begin clinical testing. The
advantage to this approach is the ability to apply a repair
directly to the annulus, where the pathologic mechanism of SUMMARY
MR is frequently located, eliminating the anatomic uncertainty The percutaneous treatment of MR is currently under investi-
regarding circumflex artery anatomy and CS proximity to the gation using a variety of devices and the initial data are fairly
annulus that plagues CS approaches. encouraging for many of them. However, none of them can
Mitralign Direct Annuloplasty System. Surgical suture currently be recommended for clinical use outside of a clinical
annuloplasty has been shown to have good long-term results, protocol. It is possible that more than one device may be
comparable to ring annuloplasty, with the possible advantage needed to properly treat MR percutaneously. However, initial
of preservation of annular contraction, usually impaired by the trials have to be conducted with one device and therefore
ring implant (22,23). Based loosely on the concept of direct patient selection will shape the future for each device. Clinical
suture annuloplasty, the Mitralign Direct Annuloplasty System trials for the percutaneous treatment of MR are particularly
[29/6/010/15:46:49] [554–567]
difficult to conduct because of several challenges. The MR 16. Palacios IF, Sanchez PL, Harrell LC, et al. Which patients benefit
population is heterogeneous in terms of etiology, limiting the from percutaneous mitral balloon valvuloplasty? Prevalvuloplasty
comparison with best medical treatment and surgery. Since and postvalvuloplasty variables that predict long-term outcome.
there are no standards for treatment and follow up reported in Circulation 2002; 105:1465–1471.
17. Perloff JK, Roberts WC. The mitral apparatus. Functional anatomy
randomized controlled fashion, investigators have to depend
of mitral regurgitation. Circulation 1972; 46:227–239.
on retrospective, single center, self reported experiences for 18. Kumanohoso T, Otsuji Y, Yoshifuku S, et al. Mechanism of higher
study design and statistical planning. Definition and selection incidence of ischemic mitral regurgitation in patients with inferior
of endpoints is equally problematic. Finally, there is no clear- myocardial infarction: quantitative analysis of left ventricular and
cut way to judge the severity of MR and the degree of is mitral valve geometry in 103 patients with prior myocardial
dependent on hemodynamic loading conditions. Therefore, infarction. J Thorac Cardiovasc Surg 2003; 125:135–143.
the efforts of all different investigators have to be applauded 19. Srichai MB, Grimm RA, Stillman AE, et al. Ischemic mitral regur-
for continuing research in this challenging field. gitation: impact of the left ventricle and mitral valve in patients
with left ventricular systolic dysfunction. Ann Thorac Surg 2005;
80:170–178.
20. Carpentier A. Cardiac valve surgery–the “French correction.”
REFERENCES J Thorac Cardiovasc Surg 1983; 86:323–337.
1. Wilkins GT, Weyman AE, Abascal VM, et al. Percutaneous bal- 21. Bonow RO, Carabello BA, Chatterjee K, et al. ACC/AHA 2006
loon dilatation of the mitral valve: an analysis of echocardio- guidelines for the management of patients with valvular heart
graphic variables related to outcome and the mechanism of disease: a report of the American College of Cardiology/American
dilatation. Br Heart J 1988; 60:299–308. Heart Association Task Force on Practice Guidelines (writing com-
2. Mueller UK, Sareli P, Essop MR. Anterior mitral leaflet retraction– mittee to revise the 1998 guidelines for the management of patients
a new echocardiographic predictor of severe mitral regurgitation with valvular heart disease) developed in collaboration with the
following balloon valvuloplasty by the Inoue technique. Am J Society of Cardiovascular Anesthesiologists endorsed by the Society
Cardiol 1998; 81:656–659. for Cardiovascular Angiography and Interventions and the Society
3. Padial LR, Abascal VM, Moreno PR, et al. Echocardiography can of Thoracic Surgeons. J Am Coll Cardiol 2006; 48:e1–e148.
predict the development of severe mitral regurgitation after 22. Nagy ZL, Bodi A, Vaszily M, et al. Five-year experience with a
percutaneous mitral valvuloplasty by the Inoue technique. Am J suture annuloplasty for mitral valve repair. Scand Cardiovasc J
Cardiol 1999; 83:1210–1213. 2000; 34:528–532.
4. Palacios IF. Farewell to surgical mitral commissurotomy for many 23. Aybek T, Risteski P, Miskovic A, et al. Seven years’ experience
patients. Circulation 1998; 97:223–226. with suture annuloplasty for mitral valve repair. J Thorac Car-
5. Ben Farhat M, Ayari M, Maatouk F, et al. Percutaneous balloon diovasc Surg 2006; 131:99–106.
versus surgical closed and open mitral commissurotomy: seven- 24. Grossi EA, Bizekis CS, LaPietra A, et al. Late results of isolated
year follow-up results of a randomized trial. Circulation 1998; mitral annuloplasty for “functional” ischemic mitral insufficiency.
97:245–250. J Card Surg 2001; 16:328–332.
6. Stefanadis CI, Stratos CG, Lambrou SG, et al. Retrograde non- 25. Bolling SF, Pagani FD, Deeb GM, et al. Intermediate-term outcome
transseptal balloon mitral valvuloplasty: immediate results and of mitral reconstruction in cardiomyopathy. J Thorac Cardiovasc
intermediate long-term outcome in 441 cases–a multicenter expe- Surg 1998; 115:381–386; discussion 7–8.
rience. J Am Coll Cardiol 1998; 32:1009–1016. 26. Bishay ES, McCarthy PM, Cosgrove DM, et al. Mitral valve sur-
7. Bahl VK, Chandra S, Jhamb DK, et al. Balloon mitral valvotomy: gery in patients with severe left ventricular dysfunction. Eur J
comparison between antegrade Inoue and retrograde non-trans- Cardiothorac Surg 2000; 17:213–221.
septal techniques. Eur Heart J 1997; 18:1765–1770. 27. Wu AH, Aaronson KD, Bolling SF, et al. Impact of mitral valve
8. Cribier A, Eltchaninoff H, Koning R, et al. Percutaneous mechanical annuloplasty on mortality risk in patients with mitral regurgita-
mitral commissurotomy with a newly designed metallic valvulo- tion and left ventricular systolic dysfunction. J Am Coll Cardiol
tome: immediate results of the initial experience in 153 patients. 2005; 45:381–387.
Circulation 1999; 99:793–799. 28. McGee EC, Gillinov AM, Blackstone EH, et al. Recurrent mitral
9. Palacios IF, Block PC, Wilkins GT, et al. Follow-up of patients regurgitation after annuloplasty for functional ischemic mitral
undergoing percutaneous mitral balloon valvotomy. Analysis of regurgitation. J Thorac Cardiovasc Surg 2004; 128:916–924.
factors determining restenosis. Circulation 1989; 79:573–579. 29. Mohty D, Orszulak TA, Schaff HV, et al. Very long-term survival
10. Hernandez R, Banuelos C, Alfonso F, et al. Long-term clinical and and durability of mitral valve repair for mitral valve prolapse.
echocardiographic follow-up after percutaneous mitral valvulo- Circulation 2001; 104:I1–I7.
plasty with the Inoue balloon. Circulation 1999; 99:1580–1586. 30. Braunberger E, Deloche A, Berrebi A, et al. Very long-term results
11. Iung B, Garbarz E, Michaud P, et al. Late results of percutaneous (more than 20 years) of valve repair with Carpentier’s techniques
mitral commissurotomy in a series of 1024 patients. Analysis of in nonrheumatic mitral valve insufficiency. Circulation 2001; 104:
late clinical deterioration: frequency, anatomic findings, and pre- I8–I11.
dictive factors. Circulation 1999; 99:3272–3278. 31. Gillinov AM, Cosgrove DM, Blackstone EH, et al. Durability of
12. Meneveau N, Schiele F, Seronde MF, et al. Predictors of event-free mitral valve repair for degenerative disease. J Thorac Cardiovasc
survival after percutaneous mitral commissurotomy. Heart 1998; Surg 1998; 116:734–743.
80:359–364. 32. Flameng W, Herijgers P, Bogaerts K. Recurrence of mitral valve
13. Lau KW, Ding ZP, Quek S, et al. Long-term (36–63 months) clinical regurgitation after mitral valve repair in degenerative valve dis-
and echocardiographic follow-up after Inoue balloon mitral com- ease. Circulation 2003; 107:1609–1613.
missurotomy. Cathet Cardiovasc Diagn 1998; 43:33–38. 33. Alkadhi H, Bettex D, Wildermuth S, et al. Dynamic cine imaging
14. Langerveld J, Thijs Plokker HW, Ernst SM, et al. Predictors of of the mitral valve with 16-MDCT: a feasibility study. Am J
clinical events or restenosis during follow-up after percutaneous Roentgenol 2005; 185:636–646.
mitral balloon valvotomy. Eur Heart J 1999; 20:519–526. 34. Fujita N, Chazouilleres AF, Hartiala JJ, et al. Quantification of
15. Kang DH, Park SW, Song JK, et al. Long-term clinical and mitral regurgitation by velocity-encoded cine nuclear magnetic
echocardiographic outcome of percutaneous mitral valvuloplasty: resonance imaging. J Am Coll Cardiol 1994; 23:951–958.
randomized comparison of Inoue and double-balloon techniques. 35. Khanna D, Miller AP, Nanda NC, et al. Transthoracic and trans-
J Am Coll Cardiol 2000; 35:169–175. esophageal echocardiographic assessment of mitral regurgitation
[29/6/010/15:46:49] [554–567]
severity: usefulness of qualitative and semiquantitative techni- percutaneous mitral annuloplasty. J Am Coll Cardiol 2004;
ques. Echocardiography 2005; 22:748–769. 44:1652–1661.
36. Maisano F, Vigano G, Blasio A, et al. Surgical isolated edge-to- 48. Kaye DM, Byrne M, Alferness C, et al. Feasibility and short-term
edge mitral valve repair without annuloplasty: clinical proof of efficacy of percutaneous mitral annular reduction for the therapy
the principle for an endovascular approach. EuroIntervention of heart failure-induced mitral regurgitation. Circulation 2003;
2006; 2:181–186. 108:1795–1797.
37. Fann JI, St Goar FG, Komtebedde J, et al. Beating heart catheter- 49. Byrne MJ, Kaye DM, Mathis M, et al. Percutaneous mitral annular
based edge-to-edge mitral valve procedure in a porcine model: reduction provides continued benefit in an ovine model of dilated
efficacy and healing response. Circulation 2004; 110:988–993. cardiomyopathy. Circulation 2004; 110:3088–3092.
38. St Goar FG, Fann JI, Komtebedde J, et al. Endovascular edge-to- 50. Webb JG, Harnek J, Munt BI, et al. Percutaneous transvenous
edge mitral valve repair: short-term results in a porcine model. mitral annuloplasty: initial human experience with device implan-
Circulation 2003; 108:1990–1993. tation in the coronary sinus. Circulation 2006; 113:851–855.
39. Feldman T, Wasserman HS, Herrmann HC, et al. Percutaneous 51. Timek TA, Dagum P, Lai DT, et al. Pathogenesis of mitral regur-
mitral valve repair using the edge-to-edge technique: six-month gitation in tachycardia-induced cardiomyopathy. Circulation
results of the EVEREST Phase I Clinical Trial. J Am Coll Cardiol 2001; 104:I47–I53.
2005; 46:2134–2140. 52. Rogers JH, Rahdert DA, Caputo GR, et al. Long-term safety and
40. The Evalve concept for mitral valve repair: results of the EVEREST durability of percutaneous septal sinus shortening (The PS(3)
trial. Available at: www.tctmd.com. Accessed November 2006. System) in an ovine model. Catheter Cardiovasc Interv 2009;
41. Dang NC, Aboodi MS, Sakaguchi T, et al. Surgical revision after 73(4):540–548.
percutaneous mitral valve repair with a clip: initial multicenter 53. Rogers JH, Macoviak JA, Rahdert DA, et al. Percutaneous septal
experience. Ann Thorac Surg 2005; 80:2338–2342. sinus shortening: a novel procedure for the treatment of functional
42. Choure AJ, Garcia MJ, Hesse B, et al. In vivo analysis of the mitral regurgitation. Circulation 2006; 113:2329–2334.
anatomical relationship of coronary sinus to mitral annulus and 54. Palacios IF, Condado JA, Brandi S, et al. First-in-man: percuta-
left circumflex coronary artery using cardiac multidetector com- neous septal sinus shortening (the PS3 system) for functional
puted tomography: implications for percutaneous coronary sinus mitral regurgitation. Am J Cardiol 2006; 98:49M.
annuloplasty. J Am Coll Cardiol 2006; 48:1938–1945. 55. Fukamachi K, Inoue M, Popovic ZB, et al. Off-pump mitral valve
43. Iansac E, Di Centa I, Al Attar N, et al. Percutaneous mitral repair using the Coapsys device: a pilot study in a pacing-induced
annuloplasty through the coronary sinus: an anatomical point of mitral regurgitation model. Ann Thorac Surg 2004; 77:688–692;
view. Circulation 2006; 114:2710. discussion 92–93.
44. Liddicoat JR, Mac Neill BD, Gillinov AM, et al. Percutaneous 56. Fukamachi K, Popovic ZB, Inoue M, et al. Changes in mitral
mitral valve repair: a feasibility study in an ovine model of acute annular and left ventricular dimensions and left ventricular pres-
ischemic mitral regurgitation. Catheter Cardiovasc Interv 2003; sure-volume relations after off-pump treatment of mitral regurgi-
60:410–416. tation with the Coapsys device. Eur J Cardiothorac Surg 2004;
45. Daimon M, Fukuda S, Adams DH, et al. Mitral valve repair with 25:352–357.
Carpentier-McCarthy-Adams IMR ETlogix annuloplasty ring for 57. Grossi EA, Saunders PC, Woo YJ, et al. Intraoperative effects of the
ischemic mitral regurgitation: early echocardiographic results Coapsys annuloplasty system in a randomized evaluation
from a multi-center study. Circulation 2006; 114:I588–I593. (RESTOR-MV) of functional ischemic mitral regurgitation. Ann
46. Gillinov AM, Liddicoat JR. Percutaneous transvenous mitral Thorac Surg 2005; 80:1706–1711.
annuloplasty. EuroIntervention 2006; 1(suppl A):A40–A43. 58. Pedersen WR, Block P, Leon M, et al. iCoapsys mitral valve repair
47. Maniu CV, Patel JB, Reuter DG, et al. Acute and chronic reduction system: percutaneous implantation in an animal model. Catheter
of functional mitral regurgitation in experimental heart failure by Cardiovasc Interv 2008;72(1):125–131.
[23/6/010/9:19:8] [568–576]
44
Hypertrophic cardiomyopathy
Matthew C. Becker, Gus Theodos, Samir R. Kapadia, and E. Murat Tuzcu
INTRODUCTION historically been utilized as first-line agents, although limited

Hypertrophic cardiomyopathy (HCM) is a unique cardiovascu- data suggest that amiodarone may reduce the risk of sudden
lar condition that may become symptomatic at any phase of life, cardiac death and improve survival in selected high-risk
from infancy to beyond 90 years of age (1–6). The hallmark of the patients with nonsustained ventricular tachycardia (11–13).
disease is myocardial hypertrophy that can be manifest in a Given that ventricular fibrillation or tachycardia is the primary
variety of recognized morphologies. Dynamic left ventricular mode of sudden cardiac death in patients with HCM, a grow-
outflow tract (LVOT) obstruction may or may not be present due ing body of data suggest that there is a significant benefit to
to asymmetric hypertrophy of the interventricular septum. With implantable cardioverter-defibrillator (ICD) therapy in patients
recent estimates suggesting that 1 in 500 of the general adult who have survived an episode of sudden cardiac death (sec-
population is affected by this autosomal dominant condition, ondary prevention) or who are at high risk for such an event
HCM is one of the most common cardiac genetic disorders (primary prevention) (Table 44.1) (5,14–16).
known, and over 12 involved genes have been identified Historically, the gold standard for the treatment of symp-
(3,5,6). The genotypic foundation of HCM is directly related to tomatic HCM has been surgical septal myotomy. Evolved from
abnormalities of the genes encoding sarcomeric proteins that the original septal myotomy first performed by Cleland (17) in
regulate the contractile, regulatory, and structural functions of the 1960s, the widely employed Morrow myectomy (18) con-
the myocardium. The consequent myocardial disarray and sists of the resection of a variable amount of myocardial tissue
hypertrophy result in a complex pathophysiologic interplay from the septum extending from the base of the aortic valve to
among LVOT obstruction, diastolic dysfunction, myocardial a region just distal to the mitral leaflets such that the LVOT is
ischemia, and mitral regurgitation. Depending on their particu- enlarged and systolic anterior motion of the mitral valve (and
lar phenotype, patients with HCM may have a wide spectrum of the resultant mitral regurgitation) is abolished (19–21). Myec-
clinical and pathologic presentations ranging from exertional tomy results in a durable reduction in outflow tract obstruction;
dyspnea, angina, palpitations, or syncope. The most fearsome a significant improvement in a patient’s functional capacity,
and dramatic complication of HCM, sudden cardiac death, is heart failure symptoms, quality of life; and may offer a life
one of the frequent causes of cardiovascular death among young expectancy similar to that of the general population (19,22–25).
people and the most common cause of mortality in competitive The operative mortality rate for the modern-day septal myec-
athletes (7). Fortunately, HCM patients at high risk for tomy is approximately 1% to 3% overall but is <1% when
sudden cardiac death constitute only a small proportion of the performed in very experienced centers (25–31).
affected population (Table 44.1). Despite the widespread avail- Introduced by Sigwart in 1995, catheter-based alcohol
ability of genetic screening tests, the diagnosis of HCM remains septal ablation has become a widely utilized alternative treat-
primarily clinical, involving the use of echocardiography to ment strategy to relieve outflow tract obstruction in symptom-
evaluate for characteristic features such as asymmetric septal atic patients who are suboptimal surgical candidates due to
hypertrophy, systolic anterior motion of the mitral valve, and comorbidities, personal preference, or who are located in areas
LVOT obstruction. without sufficient surgical expertise (32). Given the less inva-
Given the heterogeneity of the disease process, clinical sive nature and the promise of a significant reduction in
course and long-term outcomes may differ significantly in recovery time, the procedure has seen a rapid increase in
patients sharing the same mutation or even within the same popularity over the past 10 years, and is now performed
family. HCM can be a dynamic disease process that can evolve 15 to 20 times more frequently than surgical myectomy, result-
with age, and the development of left ventricular hypertrophy ing in >5000 ablations performed worldwide in total (33,34).
has been observed at all ages (8–10). Consequently, therapeutic The septal ablation technique attempts to mimic the effect of
interventions must take into consideration individual patient the more traditional Morrow myectomy by the infusion of 100%
characteristics and preferences. Accordingly, management ethanol into either the first or second septal perforator artery
strategies range from medical therapy with close outpatient supplying the septal bulge. This induces a controlled infarct in
follow-up to surgical or percutaneous remodeling of the the basal portion of the hypertrophied interventricular septum
myocardium. resulting in scarring, thinning, and akinesia. Under optimal
Medical therapy should be the initial therapeutic conditions, the result is a significant and rapid reduction in the
approach for the treatment of symptomatic patients with LVOT gradient as well as the systolic anterior motion of the
HCM. However, given the paucity of randomized trials, cur- mitral valvular apparatus (32,35–41). Limited by a paucity of
rent recommendations are based on expert opinion, clinical long-term follow-up data and the absence of randomized trials,
experience, and retrospective analyses. b-Blocking agents, vera- short-term observational studies suggest that ablation results
pamil, and disopyramide (often titrated to high doses) have in a significant reduction in LVOT gradient, a reduction in
[23/6/010/9:19:8] [568–576]
HYPERTROPHIC CARDIOMYOPATHY 569
Table 44.1 Risk Factors for Sudden Cardiac Death in Hypertrophic ANATOMIC CONSIDERATIONS
Cardiomyopathy In recognition of the fact that in current practice most patients
with HCM are diagnosed noninvasively via echocardiography
l Prior history of cardiac arrest
l
and many have not had invasive hemodynamic studies per-
Unexplained syncope (particularly if occurring with exertion)
l
formed prior to presenting for ablation, most experienced
Spontaneous sustained ventricular tachycardia
l Abnormal response (particularly blood pressure) with stress testing operators will confirm the presence of significant LVOT
l Early onset of disease obstruction by positioning an end-hole catheter in the ventric-
l Family history of cardiac arrest or sudden cardiac death ular apex and recording a slow pullback under fluoroscopy.
l Nonsustained ventricular tachycardia on Holter monitoring Alternatively, one may place a catheter in the ascending aorta
l Left ventricular thickness >30 mm and an end-hole catheter in the ventricle to provide simultane-
l Ischemia detected on perfusion testing (may be a result of ous measurement of the ascending aortic and intracavity pres-
microvascular circulation) sures. If the operator has difficulty identifying an LVOT
l Atrial fibrillation gradient under basal/resting conditions, provocation with
l Concomitant severe aortic stenosis
l
either amyl nitrate or the strain phase of the Valsalva maneuver
Concomitant congestive heart failure
l
may be attempted (51). Induction of a premature ventricular
Other comorbidities, including pulmonary embolus
and malignancy contraction with the pigtail catheter may also aid in the diag-
nosis, as it commonly results in the Brockenbrough–Braunwald
phenomenon, which refers to the diminished aortic pulse pres-
sure and increased LVOT gradient secondary to a transient
Table 44.2 Comparison of Alcohol Septal Ablation and Surgical increase in obstruction (Fig. 44.1). This maneuver can also be
Myectomy useful during the procedure, as it may isolate the appropriate
Myectomy Ablation septal perforator branch to intervene upon (52). The left ven-
Mortality <1–2% 1–2%
triculogram may have a variety of findings including systolic
Symptoms Decreased Decreased cavity obliteration, varying degrees of mitral regurgitation, and
Gradient Decreased to Decreased to occasionally the hypertrophied septum prolapsing into the
<10 mmHg <25 mmHg LVOT.
Need for pacemaker 1–2% 5–10% Standard diagnostic coronary angiography is performed
Sudden death risk Low (long term) Low (midterm) as a first step to clearly define the patient’s anatomy, identify the
Intramyocardial scar Absent Present location and size of the septal perforator branches, and evaluate
for the presence of concomitant coronary artery disease. Some-
times septal perforator branches originating from the left ante-
rior descending (LAD) artery may be compressed during systole
limiting symptoms, and improved exercise tolerance, with a due to contraction of the hypertrophied septum that contains
reported mortality equal to or less than that of surgery at 1% to them (53). Similarly, systolic compression of the LAD artery
4% (3,35,37–39,42,43). In addition, recent work suggests that resulting in a “sawfish” appearance has also been described
alcohol septal ablation may reduce psychological distress and (54). In addition to evaluating for the presence of coronary
improve feelings of well-being at three months following the artery disease, a critical component of the diagnostic angiogram
procedure (44). is proper selection of the appropriate septal perforator branch
While randomized comparative studies are lacking, cur- through which the ablation will be performed (Fig. 44.2).
rently available data suggest that septal ablation and surgical The right anterior oblique (RAO) or posterior-anterior
myectomy have similar short and mid-term success rates (PA) cranial views usually provide the best view of the septal
(Table 44.2). Postprocedurally, both modalities of septal reduc- branches as they travel through the basal interventricular
tion offer similar degrees of LVOT gradient reduction that septum and will allow proper evaluation. Often there can be
appears to be durable up to one year after either procedure substantial variation in the septal anatomy such that one sub-
(30,37,42,45–48). Furthermore, at 6- and 12-month follow-up, division runs along the left side of the septum while another
both groups were found to have similar and sustained runs along the right. The operator may gain a better under-
improvements in NYHA functional class, Canadian cardiovas- standing of this variation (i.e., on the right or left side) utilizing
cular angina class, and a reduction in the number of syncopal the left anterior oblique (LAO) projection. To reduce the like-
and presyncopal events (30,42,49). However, both procedures lihood of inducing complete heart block during ethanol infu-
have advantages as well as associated complications that fur- sion, selection of the left-sided subdivision whenever possible
ther underscore the importance of exercising the utmost care is optimal for the ablation. This can be accomplished in the
with patient selection prior to choosing either intervention. A LAO view by avoiding the subdivision of the septal branch
recent retrospective analysis of 601 patients with severely supplying the right side of the septum.
symptomatic, drug refractory HCM referred to the Mayo Clinic During angiographic assessment of the septal anatomy, a
for catheter ablation found that alcohol septal ablation offered a number of important factors must be considered with regard to
four-year survival similar to that of age-and gender-matched proper vessel selection. It is critical that the operator pay close
patients who had undergone surgical myectomy but were attention to vessel size, angulation, and distribution of myo-
noted to have a complication rate that was significantly greater cardial territories served by the given vessel. By virtue of the
than that of myectomy (50). In summary, either surgical myec- fact that larger vessels generally provide blood supply to a
tomy or alcohol ablation may be considered as a primary larger segment of myocardium, selection of septal branches
treatment modality in symptomatic patients with LVOT with a diameter >2.0 mm may result in a large infarct; this
obstruction after careful consideration of the patient’s clinical should be taken into consideration in making the decision.
situation, anatomical characteristics, and institutional expertise. Another, often underappreciated, consideration in vessel
[23/6/010/9:19:8] [568–576]
Figure 44.1 Brockenbrough–Braunwald phenomenon.
Figure 44.2 Diagnostic angiogram depicting the relationship of the first septal perforator to the plane of the aortic valve (beneath) and to the
left ventricular cavity (panels A and B). Panel C displays simultaneous injection through both the guide and pigtail catheters demonstrating the
location of the interventricular septum (white arrows and shaded segment) in relation to the left ventricular cavity. The anterior leaflet of the
mitral valve is outlined the white line and black arrows.
selection relates to the degree of angulation of the septal vessel Previous work has demonstrated that the first septal
with regard to its parent vessel. The interventionalist should artery may provide blood flow to a substantially larger region
consider that septal vessels with a high degree of angulation of myocardium than might be expected. In addition, septal
and tortuosity may be a difficult target for the delivery of the branches can exhibit marked variability in their course and
angioplasty balloon since the coronary wire may often prolapse provide blood flow to various regions of the myocardium—
back into the LAD coronary artery (51) (Fig. 44.2). including the right ventricle. Furthermore, cases in which both
Substantial anatomic variation exists with regard to the the first and second septal arteries provide blood supply to the
vessel of origin as well as the distribution of blood flow basal septum have been described (51,55). As a result, it is
supplied by the septal perforators in patients with HCM. imperative that the operator have an intimate understanding of
Most commonly noted to originate from the proximal LAD, the distribution of blood flow supplied by the selected septal
septal perforators have been observed to arise from virtually branch to accurately target the correct area for ablation and to
every major branch of the coronary tree including the left main avoid infarction of an unanticipated region or an oversized
trunk, left circumflex coronary artery, ramus intermedius, infarction of the septum itself. To better delineate the area of
diagonal branches, and, rarely, from branches of the right myocardium subtended by the selected septal vessel prior to
coronary artery (51). ethanol infusion, most experienced operators will elect to
[23/6/010/9:19:8] [568–576]
Figure 44.3 Intracardiac echocardiogram demonstrating the hypertrophied interventricular septum prior to (panel A) and after ethanol
infusion (panels B and C). Note the increased echogenicity of the basal septum following injection.
perform a selective injection of contrast under cine with the PROCEDURAL TECHNIQUE AND EQUIPMENT
concomitant use of transthoracic echocardiography utilizing Placement of a prophylactic temporary transvenous pacemaker
injectable contrast material. Intracardiac echocardiography prior to beginning the procedure is an essential initial step,
may be utilized during this procedure to better visualize the given the high incidence of complete heart block during, or in
hypertrophied septum and the course of the injected contrast the days following, ablation. To reduce the possibility of dis-
material (56) (Fig. 44.3). lodgment during the procedure or in patient transport, some
It is important to thoroughly evaluate for anatomical operators place a screw-in type pacing lead (rather than the
abnormalities that would require surgical management prior blunt-tip models) into the right ventricular apex under fluoros-
to proceeding with a catheter-based intervention. Such abnor- copy. Subsequent to placing the transvenous pacemaker, arterial
malities include anomalous papillary muscle insertion into the access is obtained using a 6- or 7-Fr short sheath, and intrave-
mitral valve, anatomically abnormal mitral valve with a long nous unfractionated heparin is administered with a target
anterior/posterior leaflet, coexistent coronary artery disease, activated clotting time of 250 to 300 seconds to prevent devel-
primary valvular disease (aortic or mitral), or subaortic mem- opment of thrombus in the guide catheters or on the wires.
brane or pannus—none of which can be adequately addressed A guiding catheter capable of providing sufficient sup-
by septal ablation (Fig. 44.4) (5,51). Furthermore, a subset of port, such as the 6- or 7-Fr XB catheter, is usually selected and
patients may be found to have an abnormal elongation and used to engage the left main coronary artery (Table 44.3). As
myomatous degeneration of the anterior mitral leaflet resulting with standard coronary interventional procedures, a 0.014-in.
in an anterior displacement of the line of coaptation with extrasupport wire with a soft tip is subsequently advanced into
resultant outflow tract obstruction. These gene-positive HCM the preselected septal perforator branch (Fig. 44.4). At this
patients have dynamic LVOT obstruction from papillary mus- point, an over-the-wire style angioplasty balloon, usually a
cle orientation independent of septal hypertrophy (57,58). 1.5 to 2 mm 8 to 12 mm, is advanced over the guidewire
These patients will require surgical consultation for consider- and into the proximal aspect of the septal vessel. Given that
ation of myectomy with plication and should not be considered septal vessels often have an acute angle at their origin from the
for catheter-based therapy (59). In addition, results of alcohol LAD, the operator may occasionally encounter difficulty in
ablation in patients with severe hypertrophy (i.e., >3.0 cm) are delivering equipment to the selected vessel. In these situations,
inconsistent, and these patients are frequently referred for stiffer guidewires are often helpful in providing the necessary
surgical correction. support for balloon placement (51). It is critical for the operator
Figure 44.4 Angiogram depicting the course of the left anterior descending artery and a septal perforator appropriate for ablation of basal-
most portion of the septum (panel A). 0.014-in. extrasupport coronary wire in the selected septal vessel with subsequent inflation of an over-
the-wire style angioplasty balloon prior to infusion of ethanol (panel B). Angiogram documenting the integrity of the left anterior descending
coronary artery and total occlusion of the recently injected septal branch (Panel C).
[23/6/010/9:19:8] [568–576]
Table 44.3 Equipment for Alcohol Septal Ablation

Equipment Specific type Comments
Guide catheter 7 Fr XB, JL4, Amplatz Provide better support
Wire 0.014-in. extrasupport wire with a soft tip Standard interventional wire, though may require stiffer wire due to
acute angulation of many septa
Balloon Over-the-wire angioplasty balloon, usually
1.5–2 mm 8–12 mm
Echocontrast 1–2 cc with balloon inflated Check placement and for reflux into left anterior descending artery
proper. Diluted 1:5 or 1:10
Alcohol 1–3 cc desiccated ethanol Smaller volumes equally efficacious
Figure 44.5 Transthoracic echocardiogram (parasternal long axis) demonstrating the grossly hypertrophied basal interventricular septum of
a patient affected by hypertrophic cardiomyopathy prior to contrast injection (panel A). Note the increased echogenicity of the basal septum
following infusion of contrast (panel B) and the echolucency immediately after ethanol infusion (panel C).
to ensure that the balloon is situated deeply enough into the solution in a 1:5 or 1:10 mixture at the time of injection. It is also
septal vessel to prevent the infused ethanol from accidentally our standard practice to utilize pulsed-wave Doppler when
refluxing into the LAD proper. However, balloon placement employing the newly diluted contrast so as to avoid destruction
too distally may result in the ethanol infusion, missing the of the microbubbles with the higher frequency continuous
basal-most portion of the septum, and result in a suboptimal wave ultrasound. Given that the basal portion of septum is
gradient reduction. responsible for the greatest extent of septal to mitral contact
Prior to proceeding further, it is critical to assess the and therefore the LVOT obstruction, the optimal septal vessel
amount and location of myocardium supplied by the selected will deliver contrast material filling to this region alone
septal vessel. This is typically accomplished through the use of (Fig. 44.5).
both echocardiographic and angiographic techniques. Follow- Given the extensive variability of the septal anatomy in
ing optimal placement, the balloon is inflated to approximately this patient population, the operator should verify that the
10 to 12 atm to occlude the vessel, and 1 to 2 cc of contrast are chosen vessel primarily supplies the proximal interventricular
slowly injected through the distal end of the balloon to assess septum and does not provide blood flow to portions of the
the full extent of myocardium supplied. While slowly infusing inferior wall (as in the presence of an occlusion of the right
the contrast material to simulate the forthcoming ethanol injec- coronary artery), left ventricular papillary musculature, or the
tion, it is imperative for the operator to confirm that the balloon right ventricular free wall (49). Should it be noted that the
occlusion is sufficient to prevent any of the infused contrast selected septal vessel provides flow to the distal septum or
from refluxing backward into the proximal LAD proper— other regions of myocardium, ethanol infusion is contraindi-
which, during the actual ethanol infusion, would result in cated as proceeding could result in infarction of an undesired
diffuse and unintended myocardial necrosis. territory or of unanticipated size. Given that it has been
Contrast echocardiography utilizing a transthoracic demonstrated that a rapid reduction in gradient can be
approach is the second commonly employed method of verify- observed with balloon occlusion of a septal perforator branch
ing the myocardial distribution subtended by a particular in many, but not all, patients, documenting a reduction in
septal vessel. Following a complete examination of the mor- LVOT gradient (generally >30%) during balloon inflation is
phology of the interventricular septum, most commonly in the another method to verify that the correct septal distribution has
parasternal short axis and apical three and four chamber views, been targeted for ablation (51) (Fig. 44.5).
1 to 2 cc of echocardiographic contrast material is infused At this point, confirmation of balloon placement and
through the balloon into the septal branch through a tubercu- appropriate function of the temporary pacemaker must again
lin-type syringe. We find that many of the currently available be verified. The balloon can migrate during the above process,
contrast agents are too concentrated, and their infusion may so another 1 to 2 cc contrast injection under fluoroscopy will
result in echocardiographic “shadowing” from the opacified ensure proper placement. Ethanol injection into the selected
ventricles. To avoid this problem, the protocol in our laboratory septal perforator may now be performed. In most cases, 1 to 3 cc
is to open the contrast vials 10 to 15 minutes prior to the time of of desiccated ethanol is sufficient—although the necessary
expected use and to further dilute the agent with a sterile saline volume can vary on the basis of the patient’s septal anatomy
[23/6/010/9:19:8] [568–576]
and the behavior of contrast during the test-infusion as per- experienced centers, the patient is then generally observed on a
formed earlier (Table 44.3) (35,37,45,51,60). If rapid washout of regular nursing floor an additional 48 to 72 hours prior to
the contrast was observed due to collateral flow, standard discharge.
practice is to reduce the volume and rate of ethanol injection Follow-up of patients after the procedure remains impor-
to ensure that other areas of myocardium are not inadvertently tant, with attention to recurrence of symptoms and arrhythmias.
damaged via these branches (51,61). Accordingly, it should be When ventricular arrhythmias are present, implantation of an
the goal of the operator to use as little ethanol as possible ICD should be considered. Regular assessment of functional
during the procedure to avoid excessive or unintended myo- capacity with exercise testing is warranted in this population.
cardial necrosis. Importantly, it has been demonstrated that For patients in whom symptoms recur and another septal per-
smaller volumes of ethanol are equally effective with regard to forator branch is available, a repeat ethanol injection can be
midterm gradient reduction while significantly reducing the performed. Prior studies have shown this to be necessary in 5%
rate of complications (62). Furthermore, the volume of alcohol to 10% of cases. If another septal branch is not present, these
used in the procedure has also been shown to be an indepen- patients are still eligible for surgical myectomy (48). For patients
dent predictor of survival—the lower, the better—in a single in which septal ablation has failed, surgical myectomy carries a
center over 10 years of follow-up (63). higher incidence of postoperative heart block versus those in
With the balloon still inflated, the ethanol is injected into whom only myectomy was performed (65).
the septal branch over a period of one to five minutes. It is
important that the operator adequately flush the catheter with
normal saline following infusion to ensure complete delivery of INDICATIONS
the previously infused ethanol into the distal vessel and, of A number of patients will continue to experience lifestyle-
great importance, to prevent reflux of infused ethanol at the modifying symptoms despite optimal medical therapy. A
time of balloon deflation. To provide maximal contact between small but definite percentage of these patients may be candi-
ethanol and myocardium, many operators will continue bal- dates for either surgical myectomy or septal alcohol ablation.
loon inflation for up to 5 to 10 additional minutes. This maneu- As has been stressed previously, it is of critical importance that
ver also helps to ensure that no reflux of ethanol into the LAD a careful screening process be in place to identify which
following balloon deflation will occur. During ethanol infusion patients are appropriate for a septal modifying procedure and
the resting LVOT gradient should be continuously monitored which type of procedure is best suited to the individual patient.
to gauge the relative success of the procedure. A successful Furthermore, it should be underscored that any form of septal
procedure is generally regarded as one in which the resting modification is not a curative, but rather symptom-reducing
LVOT gradient is reduced from >50 mmHg to <30 mmHg, or intervention. The clinical circumstances in which a surgical
in which the provocable gradient is reduced by >50% (37,51). septal-modifying procedure may be considered have been out-
The coronary guidewire is then readvanced into the lined in the updated American College of Cardiology/European
septal branch prior to disengaging the balloon to facilitate Society of Cardiology consensus statement and includes patients
removal of the balloon catheter and to maintain access to the with severe septal hypertrophy (i.e., >18 mm), resting/basal or
LAD in case of the need for reinstrumentation at a later point in inducible LVOT obstruction with gradient >50 mmHg, and
the procedure. A cineangiogram should then be performed to severely limiting heart failure symptoms (i.e., NYHA functional
document the integrity of the left main trunk and LAD. class III–IV) despite optimal medical therapy (Table 44.4)
Although total occlusion of the recently injected septal branch (4,23,25,27,31,51,66,67). Therefore, patients meeting these objec-
is most commonly observed, some phasic flow may persist in tive criteria that continue to experience significant functional
the moments following ablation (Fig. 44.4). limitation due to symptoms of chest pain, exertional dyspnea, or
Prior studies have demonstrated that patients with HCM recurrent syncope while compliant with optimal medical ther-
have a reduction in coronary flow reserve—either due to a apy and who are ineligible for surgical myectomy may be
reversible adaptive response to systolic contraction overloading considered for septal alcohol ablation.
or from a chronic remodeling process of the coronary micro- It is generally inadvisable to perform either surgical or
circulation. However, recent work suggests there is an imme- percutaneous septal modification in minimally symptomatic
diate improvement in coronary flow reserve following alcohol patients with obstructive HCM, given the short-term risks
septal ablation suggesting that the dynamic response to the inherent to both procedures and the paucity of data suggesting
pressure overload from the LVOT obstruction is a critical that the long-term outcome of untreated patients is worse than
component of this pathophysiology (64). those undergoing the procedure. Furthermore, as outlined ear-
Given the potential for serious intra- and periprocedural lier, both forms of septal reduction may potentially be associ-
complications, observation of the patient in a coronary inten- ated with serious complications—the risk of which may well
sive care unit experienced with the nuances of the postproce- outweigh any possible benefit in a minimally symptomatic
dural care of such patients is advisable for 24 to 48 hours.
Measured serum levels of creatinine phosphokinase (CPK) Table 44.4 Selection Criteria for Alcohol Septal Ablation
often reach levels between 800 and 1200 U/L following ethanol
l NYHA heart failure symptoms (class III–IV) despite maximal
injection, though variation in the observed value is dependent
on caliber of the injected vessel, volume of ethanol applied, and medical therapy
l Left ventricular outflow tract gradient >30 mmHg at rest or
assays used (4,34,35,45,62). If telemetry monitoring reveals no
>50 mmHg with provocation
episodes of heart block or bradyarrhythmias after 24 to l Absence of mitral valve or papillary muscle abnormalities
48 hours, the temporary pacemaker can safely be removed. l Septal thickness >16–18 mm
Development of complete heart block in patients with no under- l Compatible anatomy of septal branches
lying conduction abnormalities is rare, but if it does occur, l Absence of significant coronary artery disease
the presence of a temporary pacemaker is lifesaving. In most
[23/6/010/9:19:8] [568–576]
patient. In addition, it should be stressed that these procedures patients following septal ablation (4,36,37,51). The incidence of
should be performed at experienced centers as there is a complete heart block exists and ranges widely across the liter-
significant learning curve that impacts the odds of a successful ature (anywhere from 5–40%), with an average value of 12% to
outcome and the amelioration of symptoms (68). Finally, it is 15% at high-volume centers (4,30,37,39,51,69). High-degree
important to emphasize that at present there is no clear data atrioventricular block requiring permanent pacemaker implan-
suggesting that relief of the outflow tract gradient through tation following the procedure has been correlated with pre-
septal modification will positively alter a patient’s overall risk existing left bundle branch block and rapid injection of ethanol
of progressing to the end-stage disease state. during the procedure (69). As discussed previously, special
care must be taken to ensure no ethanol spills into the LAD.
When this occurs, it is a catastrophic event, often resulting in a
CLINICAL ASPECTS large mid- to distal anterior wall myocardial infarction, with an
It is of critical importance that the clinicians thoroughly eval- associated increase in mortality. Rare cases of coronary dissec-
uate the patient for anatomical abnormalities that would tion related to use of an extrasupport guidewire or guide
require surgical management prior to proceeding with a cath- catheter trauma have been reported in the literature. Cardiac
eter-based intervention. These abnormalities would include tamponade is another rare complication most commonly the
anomalous papillary muscle insertion into the mitral valve, result of perforation through the right ventricular apex during
anatomically abnormal mitral valve with a long anterior/poste- temporary pacemaker insertion or during interatrial septal
rior leaflet, coexistent coronary artery disease, primary valvular puncture for hemodynamic monitoring. Another infrequent
disease (aortic or mitral), or subaortic membrane or pannus— complication following the ablation is ventricular septal rup-
none of which can be adequately addressed by septal ablation ture, which can occur when an excess of ethanol is injected,
(4,51). In addition, outflow tract obstruction resulting from resulting in an extensive area of infarction (51). Ventricular
abnormal elongation and/or myomatous degeneration of the arrhythmias following the procedure are rare in the first
anterior mitral leaflet is a rare but important discovery that 48 hours and can usually be managed conservatively while
would not be readily amenable to a percutaneous solution. on telemetry observation. It remains a source of concern
Therefore, these patients will require surgical consultation for whether, due to the intramyocardial scar that develops follow-
consideration of myectomy with plication and should not be ing ablation, patients (particularly younger ones) may be at risk
considered for catheter-based therapy (Table 44.5) (59). for late ventricular arrhythmias. However, this has so far not
A thorough patient history including a detailed assess- been convincingly demonstrated in the literature (4,37,60,70).
ment of functional status is essential for appropriate patient A recent prospective study has demonstrated a reduced num-
selection. Regardless of the magnitude of the resting or ber of ICD discharges in patients with HCM following
dynamic LVOT gradient it is inadvisable to perform either alcohol septal ablation, arguing against the assumption that
surgical or percutaneous septal modification in patients with- the procedure is proarrhythmic (71).
out significant symptoms (or symptoms that are easily con-
trolled with medications), given the short-term risks inherent to
these procedures and the paucity of data suggesting that the SPECIAL ISSUES/CONSIDERATIONS
long-term outcome of untreated patients is worse than those As with any specialized procedure, local expertise must be
undergoing the procedure. Therefore, given that both forms of considered when deciding between surgical myectomy and
septal reduction can have potentially serious complications, the alcohol septal ablation. To treat this condition successfully, the
risk of either form of procedure may well outweigh any pos- interventionalist should not only be technically skilled but also
sible benefit in a minimally symptomatic patient. ensure appropriate clinical follow-up and management post-
procedure. Each patient is unique, and his/her individual anat-
omy, symptoms, and risk of surgery will impact the decision to
LIMITATIONS perform alcohol septal ablation versus surgical myectomy.
Complications following ethanol septal ablation are infrequent Advances in MRI have improved diagnosis of the condi-
and comparable to those following myectomy. Although a left tion and give valuable information regarding location of pap-
bundle branch block is often observed following septal myec- illary muscle insertion. Local expertise in this field is therefore
tomy, a right bundle branch block is observed in up to 80% of also important. Given the risk of high-degree atrioventricular
block and potential need for permanent pacemaker implanta-
tion, centers with adequate electrophysiology support are also
preferable. Recall that right bundle branch block is more com-
Table 44.5 Conditions in Which Surgical Correction of HCM
mon with alcohol septal ablation, which differs from surgical
Should Be Considered.
myectomy, in which left bundle branch block is more common.
l Coexistent coronary artery disease not amenable to When surgical myectomy must be performed following failed
percutaneous intervention alcohol septal ablation, the risk of requiring permanent pace-
l Primary valvular disease (aortic or mitral) maker increases further.
l Anomalous papillary muscle insertion into the mitral valve
l Anatomically abnormal mitral valve with a long anterior/posterior
leaflet
l Subaortic membrane or pannus
CONCLUSIONS
l Abnormal elongation and myomatous degeneration of the HCM is a unique cardiovascular condition associated with
anterior mitral leaflet (resulting in an anterior displacement of the many different genetic mutations that result in a variety of
line of coaptation with resultant outflow tract obstruction) phenotypes and variable clinical presentations at any age.
l Patients with severe hypertrophy (i.e., >3.0 cm) Present in a minority of this patient population, outflow tract
obstruction may vary significantly depending on physiologic
[23/6/010/9:19:8] [568–576]
loading conditions, adrenergic state, and specific medications. 15. Epstein AE, DiMarco JP, Ellenbogen KA, et al. ACC/AHA/HRS
Despite rapid advances in genetic screening, imaging modal- 2008 Guidelines for Device-Based Therapy of Cardiac Rhythm
ities such as echocardiography and MRI remain the primary Abnormalities: a report of the American College of Cardiology/
methods of diagnosis. Given the substantial variation in this American Heart Association Task Force on Practice Guidelines
(Writing Committee to Revise the ACC/AHA/NASPE 2002
population of patients, therapeutic interventions must take into
Guideline Update for Implantation of Cardiac Pacemakers and
consideration individual patient characteristics and preferen- Antiarrhythmia Devices) developed in collaboration with the
ces. Accordingly, management strategies range from medical American Association for Thoracic Surgery and Society of Tho-
therapy with close outpatient follow-up to surgical or percuta- racic Surgeons. J Am Coll Cardiol 2008; 51:e1–e62.
neous remodeling of the myocardium. While the initial inter- 16. Maron BJ, Spirito P, Shen WK, et al. Implantable cardioverter-
vention for symptomatic patients should be medical therapy, defibrillators and prevention of sudden cardiac death in hyper-
patients at high risk for sudden cardiac death should be con- trophic cardiomyopathy. JAMA 2007; 298:405–412.
sidered for ICD therapy. In the setting of symptoms refractory 17. Cleland W. The surgical management of obstructive cardiomyop-
to optimal medical management, surgical myectomy continues athy. J Cardiovasc Surg (Torino) 1968; 4:489–491.
to be the gold standard therapy for HCM in appropriately risk- 18. Morrow AG. Hypertrophic subaortic stenosis. Operative methods
utilized to relieve left ventricular outflow obstruction. J Thorac
stratified patients in areas where an experienced team and
Cardiovasc Surg 1978; 76:423–430.
facility are available. However, when surgical therapy is not 19. Maron BJ, Dearani JA, Ommen SR, et al. The case for surgery in
feasible or possible, alcohol septal ablation provides an excel- obstructive hypertrophic cardiomyopathy. J Am Coll Cardiol
lent alternative treatment modality when performed in appro- 2004; 44:2044–2053.
priately selected patients in an experienced center. 20. Maron BJ, Harding AM, Spirito P, et al. Systolic anterior motion of
the posterior mitral leaflet: a previously unrecognized cause of
dynamic subaortic obstruction in patients with hypertrophic
REFERENCES cardiomyopathy. Circulation 1983; 68:282–293.
1. Braunwald E, Lambrew CT, Rockoff SD, et al. Idiopathic Hypertro- 21. Spirito P, Maron BJ. Significance of left ventricular outflow tract
phic Subaortic Stenosis. I. A Description of the Disease Based Upon cross-sectional area in hypertrophic cardiomyopathy: a two-dimen-
an Analysis of 64 Patients. Circulation 1964; 30:(suppl 4):3–119. sional echocardiographic assessment. Circulation 1983; 67:1100–1108.
2. Cecchi F, Olivotto I, Montereggi A, et al. Hypertrophic cardiomy- 22. Krajcer Z, Leachman RD, Cooley DA, et al. Mitral valve replace-
opathy in Tuscany: clinical course and outcome in an unselected ment and septal myomectomy in hypertrophic cardiomyopathy.
regional population. J Am Coll Cardiol 1995; 26:1529–1536. Ten-year follow-up in 80 patients. Circulation 1988; 78:I35–I43.
3. Maron BJ. Hypertrophic cardiomyopathy: a systematic review. 23. Maron BJ, Merrill WH, Freier PA, et al. Long-term clinical course
JAMA 2002; 287:1308–1320. and symptomatic status of patients after operation for hypertro-
4. Maron BJ, McKenna WJ, Danielson GK, et al. American College of phic subaortic stenosis. Circulation 1978; 57:1205–1213.
Cardiology/European Society of Cardiology clinical expert con- 24. Ommen SR. The effect of surgical myectomy on survival of
sensus document on hypertrophic cardiomyopathy. A report of patients with hypertrophic cardiomyopathy. J Am Coll Cardiol
the American College of Cardiology Foundation Task Force on 2004; 43:215A.
Clinical Expert Consensus Documents and the European Society 25. Spirito P, Seidman CE, McKenna WJ, et al. The management of
of Cardiology Committee for Practice Guidelines. J Am Coll hypertrophic cardiomyopathy. N Engl J Med 1997; 336:775–785.
Cardiol 2003; 42:1687–1713. 26. Heric B, Lytle BW, Miller DP, et al. Surgical management of
5. Maron BJ, McKenna WJ, Elliott P, et al. Hypertrophic cardiomy- hypertrophic obstructive cardiomyopathy. Early and late results.
opathy. JAMA 1999; 282:2302–2303. J Thorac Cardiovasc Surg 1995; 110:195–206; discussion –206–208.
6. Spirito P, Chiarella F, Carratino L, et al. Clinical course and 27. McCully RB, Nishimura RA, Tajik AJ, et al. Extent of clinical
prognosis of hypertrophic cardiomyopathy in an outpatient pop- improvement after surgical treatment of hypertrophic obstructive
ulation. N Engl J Med 1989; 320:749–755. cardiomyopathy. Circulation 1996; 94:467–471.
7. Maron BJ, Shirani J, Poliac LC, et al. Sudden death in young 28. Merrill WH, Friesinger GC, Graham TP Jr., et al. Long-lasting
competitive athletes. Clinical, demographic, and pathological improvement after septal myectomy for hypertrophic obstructive
profiles. JAMA 1996; 276:199–204. cardiomyopathy. Ann Thorac Surg 2000; 69:1732–1735; discussion 5–6.
8. Lewis JF, Maron BJ. Clinical and morphologic expression of 29. Mohr R, Schaff HV, Danielson GK, et al. The outcome of surgical
hypertrophic cardiomyopathy in patients > or ¼ 65 years of treatment of hypertrophic obstructive cardiomyopathy. Experi-
age. Am J Cardiol 1994; 73:1105–1111. ence over 15 years. J Thorac Cardiovasc Surg 1989; 97:666–674.
9. Olivotto I, Maron MS, Adabag AS, et al. Gender-related differ- 30. Qin JX, Shiota T, Lever HM, et al. Outcome of patients with
ences in the clinical presentation and outcome of hypertrophic hypertrophic obstructive cardiomyopathy after percutaneous
cardiomyopathy. J Am Coll Cardiol 2005; 46:480–487. transluminal septal myocardial ablation and septal myectomy
10. Spirito P, Maron BJ. Absence of progression of left ventricular surgery. J Am Coll Cardiol 2001; 38:1994–2000.
hypertrophy in adult patients with hypertrophic cardiomyopathy. 31. Schulte HD, Borisov K, Gams E, et al. Management of symptom-
J Am Coll Cardiol 1987; 9:1013–1017. atic hypertrophic obstructive cardiomyopathy–long-term results
11. McKenna WJ. Does treatment influence the natural history of after surgical therapy. Thorac Cardiovasc Surg 1999; 47:213–218.
patients with hypertrophic cardiomyopathy? Drugs 1985; 32. Sigwart U. Non-surgical myocardial reduction for hypertrophic
29(suppl 3): 53–56. obstructive cardiomyopathy. Lancet 1995; 346:211–214.
12. McKenna WJ, Kleinebenne A. Arrhythmias in hypertrophic car- 33. Maron BJ. Role of alcohol septal ablation in treatment of obstruc-
diomyopathy. Significance and therapeutic consequences. Herz tive hypertrophic cardiomyopathy. Lancet 2000; 355:425–426.
1985; 10:91–101. 34. Roberts R, Sigwart U. Current concepts of the pathogenesis and
13. McKenna WJ, Oakley CM, Krikler DM, et al. Improved survival treatment of hypertrophic cardiomyopathy. Circulation 2005;
with amiodarone in patients with hypertrophic cardiomyopathy 112:293–296.
and ventricular tachycardia. Br Heart J 1985; 53:412–416. 35. Faber L, Meissner A, Ziemssen P, et al. Percutaneous transluminal
14. Boriani G, Maron BJ, Shen WK, et al. Prevention of sudden death septal myocardial ablation for hypertrophic obstructive cardiomy-
in hypertrophic cardiomyopathy: but which defibrillator for opathy: long term follow up of the first series of 25 patients. Heart
which patient? Circulation 2004; 110:e438–e442. 2000; 83:326–331.
[23/6/010/9:19:8] [568–576]
36. Flores-Ramirez R, Lakkis NM, Middleton KJ, et al. Echocardio- angiographic sign of hypertrophic cardiomyopathy. Circulation
graphic insights into the mechanisms of relief of left ventricular 1982; 66:800–803.
outflow tract obstruction after nonsurgical septal reduction ther- 55. Singh M, Edwards WD, Holmes DR Jr., et al. Anatomy of the first
apy in patients with hypertrophic obstructive cardiomyopathy. septal perforating artery: a study with implications for ablation
J Am Coll Cardiol 2001; 37:208–214. therapy for hypertrophic cardiomyopathy. Mayo Clin Proc 2001;
37. Gietzen FH, Leuner CJ, Raute-Kreinsen U, et al. Acute and long- 76:799–802.
term results after transcoronary ablation of septal hypertrophy 56. Alfonso F, Martin D, Fernandez-Vazquez F. Intracardiac echocar-
(TASH). Catheter interventional treatment for hypertrophic diography guidance for alcohol septal ablation in hypertrophic
obstructive cardiomyopathy. Eur Heart J 1999; 20:1342–1354. obstructive cardiomyopathy. J Invasive Cardiol 2007; 19:E134–
38. Knight C, Kurbaan AS, Seggewiss H, et al. Nonsurgical septal E136.
reduction for hypertrophic obstructive cardiomyopathy: outcome 57. Austin BA, Kwon DH, Smedira NG, et al. Abnormally thickened
in the first series of patients. Circulation 1997; 95:2075–2081. papillary muscle resulting in dynamic left ventricular outflow
39. Lakkis NM, Nagueh SF, Dunn JK, et al. Nonsurgical septal reduc- tract obstruction: an unusual presentation of hypertrophic cardi-
tion therapy for hypertrophic obstructive cardiomyopathy: one- omyopathy. J Am Soc Echocardiogr 2009; 22:105 e5–e6.
year follow-up. J Am Coll Cardiol 2000; 36:852–855. 58. Kwon DH, Setser RM, Thamilarasan M, et al. Abnormal papillary
40. Lakkis NM, Nagueh SF, Kleiman NS, et al. Echocardiography- muscle morphology is independently associated with increased
guided ethanol septal reduction for hypertrophic obstructive left ventricular outflow tract obstruction in hypertrophic cardio-
cardiomyopathy. Circulation 1998; 98:1750–1755. myopathy. Heart 2008; 94:1295–1301.
41. Spirito P, Seidman CE, McKenna WJ, et al. The management of 59. Klues HG, Maron BJ, Dollar AL, et al. Diversity of structural
hypertrophic cardiomyopathy. N Engl J Med 1997; 336:775–785. mitral valve alterations in hypertrophic cardiomyopathy. Circula-
42. Firoozi S, Elliott PM, Sharma S, et al. Septal myotomy-myectomy tion 1992; 85:1651–1660.
and transcoronary septal alcohol ablation in hypertrophic obstruc- 60. Kuhn H, Gietzen FH, Leuner C, et al. Transcoronary ablation of
tive cardiomyopathy. A comparison of clinical, haemodynamic septal hypertrophy (TASH): a new treatment option for hypertro-
and exercise outcomes. Eur Heart J 2002; 23:1617–1624. phic obstructive cardiomyopathy. Z Kardiol 2000; 89(suppl 4):
43. Ruzyllo W, Chojnowska L, Demkow M, et al. Left ventricular IV41–IV54.
outflow tract gradient decrease with non-surgical myocardial 61. Faber L, Seggewiss H, Gleichmann U. Percutaneous transluminal
reduction improves exercise capacity in patients with hypertro- septal myocardial ablation in hypertrophic obstructive cardiomy-
phic obstructive cardiomyopathy. Eur Heart J 2000; 21:770–777. opathy: results with respect to intraprocedural myocardial con-
44. Serber ER, Sears SF, Nielsen CD, et al. Depression, anxiety, and trast echocardiography. Circulation 1998; 98:2415–2421.
quality of life in patients with obstructive hypertrophic cardiomy- 62. Veselka J, Duchonova R, Prochazkova S, et al. Effects of varying
opathy three months after alcohol septal ablation. Am J Cardiol ethanol dosing in percutaneous septal ablation for obstructive
2007; 100:1592–1597. hypertrophic cardiomyopathy on early hemodynamic changes.
45. Boekstegers P, Steinbigler P, Molnar A, et al. Pressure-guided Am J Cardiol 2005; 95:675–678.
nonsurgical myocardial reduction induced by small septal infarc- 63. Kuhn H, Lawrenz T, Lieder F, et al. Survival after transcoronary
tions in hypertrophic obstructive cardiomyopathy. J Am Coll ablation of septal hypertrophy in hypertrophic obstructive cardi-
Cardiol 2001; 38:846–853. omyopathy (TASH): a 10-year experience. Clin Res Cardiol 2008;
46. van Dockum WG, Beek AM, ten Cate FJ, et al. Early onset and 97:234–243.
progression of left ventricular remodeling after alcohol septal 64. Jaber WA, Yang EH, Nishimura RA, et al. Immediate improvement
ablation in hypertrophic obstructive cardiomyopathy. Circulation in coronary flow reserve after alcohol septal ablation in patients
2005; 111:2503–2508. with hypertrophic obstructive cardiomyopathy. Heart 2008.
47. Nagueh SF, Ommen SR, Lakkis NM, et al. Comparison of ethanol 65. Nagueh SF, Buergler JM, Quinones MA, et al. Outcome of surgical
septal reduction therapy with surgical myectomy for the treat- myectomy after unsuccessful alcohol septal ablation for the treat-
ment of hypertrophic obstructive cardiomyopathy. J Am Coll ment of patients with hypertrophic obstructive cardiomyopathy.
Cardiol 2001; 38:1701–1706. J Am Coll Cardiol 2007; 50:795–798.
48. Ralph-Edwards A, Woo A, McCrindle BW, et al. Hypertrophic 66. Maron MS, Olivotto I, Betocchi S, et al. Effect of left ventricular
obstructive cardiomyopathy: comparison of outcomes after myec- outflow tract obstruction on clinical outcome in hypertrophic
tomy or alcohol ablation adjusted by propensity score. J Thorac cardiomyopathy. N Engl J Med 2003; 348:295–303.
Cardiovasc Surg 2005; 129:351–358. 67. Robbins RC, Stinson EB. Long-term results of left ventricular
49. Nagueh SF, Lakkis NM, He ZX, et al. Role of myocardial contrast myotomy and myectomy for obstructive hypertrophic cardiomy-
echocardiography during nonsurgical septal reduction therapy for opathy. J Thorac Cardiovasc Surg 1996; 111:586–594.
hypertrophic obstructive cardiomyopathy. J Am Coll Cardiol 68. van der Lee C, Scholzel B, ten Berg JM, et al. Usefulness of clinical,
1998; 32:225–229. echocardiographic, and procedural characteristics to predict out-
50. Sorajja P, Valeti U, Nishimura RA, et al. Outcome of alcohol septal come after percutaneous transluminal septal myocardial ablation.
ablation for obstructive hypertrophic cardiomyopathy. Circulation Am J Cardiol 2008; 101:1315–1320.
2008; 118:131–139. 69. Chang SM, Nagueh SF, Spencer WH III, et al. Complete heart
51. Holmes DR Jr., Valeti US, Nishimura RA. Alcohol septal ablation block: determinants and clinical impact in patients with hypertro-
for hypertrophic cardiomyopathy: indications and technique. phic obstructive cardiomyopathy undergoing nonsurgical septal
Catheter Cardiovasc Interv 2005; 66:375–389. reduction therapy. J Am Coll Cardiol 2003; 42:296–300.
52. Soon CY, Buergler JM. Alcohol septal ablation and the Brocken- 70. Maron BJ, Shen WK, Link MS, et al. Efficacy of implantable
brough-Braunwald phenomenon. Catheter Cardiovasc Interv cardioverter-defibrillators for the prevention of sudden death in
2008; 72:1016–1024. patients with hypertrophic cardiomyopathy. N Engl J Med 2000;
53. Pichard AD, Meller J, Teichholz LE, et al. Septal perforator com- 342:365–373.
pression (narrowing) in idiopathic hypertrophic subaortic steno- 71. Cuoco FA, Spencer WH III, Fernandes VL, et al. Implantable
sis. Am J Cardiol 1977; 40:310–314. cardioverter-defibrillator therapy for primary prevention of sud-
54. Brugada P, Bar FW, de Zwaan C, et al. “Sawfish” systolic den death after alcohol septal ablation of hypertrophic cardiomy-
narrowing of the left anterior descending coronary artery: an opathy. J Am Coll Cardiol 2008; 52:1718–1723.
[18/6/010/21:42:51] [577–583]
45
Atrial appendage exclusion

David R. Holmes, Jr. and Rebecca Fountain
INTRODUCTION ticulum that arises from the left atrium and lies within the
Stroke is the most feared complication of cardiovascular disease pericardium superior to the left ventricular free wall and next to
and occurs with an incidence of approximately 780,000 cases the superior and lateral aspect of the main pulmonary artery. It
per year in the United States (1). Brain ischemia is the most varies significantly in structure, volume, and three-dimensional
common cause, occurring in almost 90% of cases. Stroke is the configuration (Fig. 45.2).
third most common cause of mortality and a leading cause of Veinot et al. (33) evaluated 500 normal autopsy hearts.
disability and accounts for approximately $60 billion in annual This series included 25 male and 25 female subjects from each
costs. The impact of stroke on a patient and family cannot be decade of life for a total of 10 decades. A striking finding was
overemphasized because it brings with it the threat of loss of that 54% of specimens had two lobes of the left atrial append-
independence and irreparable loss of function. Accordingly, age while only 20% of specimens had a single lobe. These lobes
because of these issues, attempts at stroke prevention have were found to lie in multiple planes and have a variable
received considerable emphasis. distance from ostium to the distal tip. More recent data from
Atrial fibrillation is the most common sustained cardiac the SPARC study also found that approximately 50% of
arrhythmia and is increasing in frequency as the population patients have multiple lobes (34,35). In addition to wide vari-
ages (2–11). In individuals who are over 40 years of age, the ability in number of lobes, the volume and dimensions also
lifetime risk of atrial fibrillation is one in four. The relationship vary. In a study of 220 cases (34), the left atrial appendage
between increasing age, atrial fibrillation, and stroke has been volume varied widely up to 19.2 mL and the length ranged
studied intensively. Stroke occurs at an annual rate of 5% in from 16 to 51 mm.
patients with atrial fibrillation; this rate increases with increas- This variability in dimensions has significant implica-
ing age of the patients. In patients >80 years of age, atrial tions for approaches aimed at left atrial appendage exclusion.
fibrillation is the cause of approximately 25% of all strokes. As The approach with the best chance to most completely and
the population ages and atrial fibrillation increases, there will reliably abolish stroke will require the ability to fully cover the
be increased need for strategies to prevent stroke. ostium of all lobes as well as the origin of the left atrial
Although multiple randomized trials have documented appendage. Exclusion of that portion of the left atrial append-
the effectiveness of warfarin therapy for stroke prevention in the age that contains pectinate muscles may also be important as
setting of atrial fibrillation compared with placebo, aspirin alone, thrombus could potentially develop between adjacent muscles.
or aspirin in combination with clopidogrel, many patients are Residual flow into a patent lobe may still result in an increased
not treated with warfarin (12–27). This is a result of the narrow incidence of stroke. The length of the left atrial appendage and
therapeutic index associated with warfarin, patient compliance, its angulation is also important, as any implantable device must
the presence of absolute or relative contraindications particularly be able to fit into the body of the left atrial appendage without
related to the potential for bleeding, and the wide variability in protruding out into the body of the left atrium itself.
dosing schedules that make treatment difficult and inconvenient. Assessment of the specific details of left atrial appendage
These issues form the basis for the fact that only approximately anatomy is essential for patient selection and procedural per-
50% of high-risk patients with atrial fibrillation are treated with formance. Transesophageal echocardiogram (TEE) is the most
warfarin (27). Although new agents are being evaluated, the commonly used approach but computed tomography (CT) is
lifetime potential for costs, side effects, and compliance, as well used with increasing frequency (Fig. 45.3).
as side effects make these less attractive (28,29). There are several essential pieces of information:
Multiple sources of data including pathological and Identifying the presence or absence of thrombus. If a thrombus
echocardiographic studies have documented that the left atrial is present as a filling defect or is suspected because of the
appendage is the site of thrombus formation in approximately presence of a dense cloud of echoes (smoke), then percutaneous
90% of patients with nonvalvular atrial fibrillation (Fig. 45.1) approaches and perhaps all nonsurgical approaches should be
(30). This finding has led to the development of several avoided because of the potential for embolization during the
approaches for exclusion of the left atrial appendage and one procedure. In this setting, warfarin should be administered and
randomized trial. The intent of these approaches has been to repeat imaging performed to document either persistence or
prevent stroke without the need for warfarin anticoagulation. disappearance of the thrombus. If the thrombus is found to
have disappeared, then device implantation can be planned
ANATOMIC CONSIDERATIONS and performed.
Consideration of the application and potential role of these Dimensions of the left atrial appendage and relationship to the
approaches requires understanding of the anatomy of the left pulmonary veins. Budge et al. (31) compared left atrial append-
atrial appendage (31–33). The left atrial appendage is a diver- age morphology in patients with atrial fibrillation using TEE,
[18/6/010/21:42:51] [577–583]
Table 45.1 Comparison of Left Atrial Appendage Anatomy Depend-

ing on Specific Imaging Modality in 53 Patients with Atrial Fibrillation
TEE (SD) CTp (SD) CTsg (SD)
No. of LAA lobes 1.21 0.41 1.11 0.32 1.08 0.27
LAA os—short 20 3.7
axis (mm)
LAA os—long 26.1 6.4 26.2 4.1 28.5 4.5
axis (mm)
LAA depth (mm) 35.9 6.2 25.1 6.3 29.2 5.7
LAA width (mm) 23.0 5.8 23.5 4.2
LAA volume (mL) 9.8 4.2
LAA volume (mL) 111.6 30.4
Abbreviations: LAA, left atrial appendage; LA, left atrium; SD, stan-
dard deviation; TEE, transesophageal echocardiography; CTp, planar
computed tomography; CTsg, segmented computed tomography.
planar CT, and segmented three-dimensional CT in 53 patients.

Figure 45.1 Echocardiography documenting thrombus within the They measured maximal left atrial appendage orifice diameter,
left atrial appendage. width, and depth with each modality. In addition, they assessed
the relationship of the left atrial appendage with the left pulmo-
nary veins. There were quantitative differences in measurements
depending on which modality was used (Table 45.1). In general,
the orifice diameter values measured by TEE and planar CT
were similar while values obtained using segmented reconstruc-
tion were larger. This has important implications for device
sizing, because with the current device, there are multiple dif-
ferent sizes depending on the orifice diameter. There were also
differences in the mean left atrial appendage depth which
ranged from 25.1 mm with planar CT to 35.9 mm with TEE.
This also has important implications for device selection and
size. Early generation devices were relatively long; more recent
devices are shorter and can fit in most anatomic situations. Of
importance is the fact that in the Budge et al. (31) study, in
contrast to the pathology series, the majority of appendages
Figure 45.2 Casts of the left atrial appendage documenting vari- studied were single lobed. The presence of multiple lobes can
ability in shape and size.
compound the issue of device placement. In some patients, not
all lobes can be covered, while in others, to cover the ostium
adequately, one of the lobes must be entered deeply for device
placement. If the angulation between lobes is large, this can
create problems trying to achieve a stable position.
The relationship between the left atrial appendage and pulmo-
nary veins. Budge et al. (31) found that the superior aspect of the
plane of the left atrial appendage was typically located in
the mid portion of the left superior pulmonary vein while the
inferior aspect was high relative to the plane of the left inferior
pulmonary vein. These relationships may have important
implications for accessing the left atrial appendage as well as
for device placement. Depending on the amount of protrusion
into the left atrium by a device, it is possible to impinge on the
orifice of one of the left pulmonary veins thereby impeding
normal flow.
APPROACHES AND OUTCOME

Although several devices have been evaluated for left atrial
appendage occlusion (36–44), the single randomized trial
Figure 45.3 CT imaging documenting the relationship between the involves the WatchmanTM device (40) (Fig. 45.4), which is a
left atrial appendage (LAA), left atrium (LA), mitral valve (MV), and self-expanding nitinol frame structure with fixation barbs and a
left ventricle (LV). Note the angulation into the LA, which may permeable polyethylene membrane (39,40,42,43). This device is
complicate device placement. available in diameters ranging from 21 to 33 mm to accommo-
date variations in left atrial appendage anatomy.
[18/6/010/21:42:51] [577–583]
ATRIAL APPENDAGE EXCLUSION 579
Approaches to the trans-septal catheterization vary con-

siderably. The goal is to cross the atrial septum at the level of
the fossa ovalis. In many laboratories, biplane fluoroscopic
imaging is used with orthogonal views either AP and lateral
or 308 RAO/608 LAO. The most important complication to
avoid is entry into the ascending aorta. While entry with the
Brockenborough needle into the aorta does not necessarily lead
to catastrophic bleeding, if high pressure is identified with
needle puncture, it is essential to avoid placement of a larger
dilator and sheath, because that results in major hemorrhage.
More recently, intracardiac ultrasound has been used
more extensively for trans-septal catheterization. With this
approach, the needle can be visualized within the right atrium
and the details of septal anatomy can be characterized. In some
Figure 45.4 The Watchman (Atritech, Inc., Plymouth, Minnesota, patients, particularly those who have had prior trans-septal
U.S.) implant is a composite device with a self-expanding nitinol procedures, the atrial septum is fibrotic and the needle may
frame with a permeable polyester fabric on the atrial facing surface slide cranial leading to a puncture site that is too superior.
of the device. Fixation barbs help to anchor the device in position. Similarly, a fatty limbus may impact on the entrance site. With
intracardiac ultrasound, the needle can be visualized as it
enters the left atrium using injection of saline or contrast. The
level of puncture of the intra atrial septum is of importance
because it impacts on the ability to optimize a guiding sheath
into the left atrial appendage. If the left atrial appendage
angulates quickly caudally, then a lower septal puncture
helps so that the guiding sheath can be placed more coaxially
into the appendage. The development and subsequent wide
use of steerable guiding sheaths will help to minimize this
issue.
Once safe access to the left atrium is accomplished,
heparin is administered. Typically, an activated clotting time
is measured with a goal of approximately 250 seconds. It is
important to administer heparin promptly because thrombus
can form in the long sheaths and can embolize during the
Figure 45.5 Experimental animal preparation documenting place- procedure.
ment of the Watchman (Atritech, Inc., Plymouth, Minnesota, U.S.) The goal of the procedure is safe access onto the left atrial
device at the origin of the left atrial appendage (LAA) (right) and appendage once a sheath has been placed into the left atrium.
endothelialization of the atrial surface (left). Knowledge of the detailed anatomy is important, particularly
the relationship between the left atrial appendage and pulmo-
nary veins. The large 14-Fr sheath for device delivery can easily
damage the left atrial wall particularly the dome, or the left
The device is implanted using a trans-septal approach atrial appendage itself. Although several approaches can be
and a catheter-based delivery system to cover the ostium of the used, the current recommended approach is to engage the left
left atrial appendage (Fig. 45.5). Device implantation is guided atrial appendage with a pigtail catheter. This approach mini-
by fluoroscopy as well as TEE to verify proper position and mizes the potential for trauma that can result in perforation of
stability. the left atrial appendage. Once the left atrial appendage is
entered with the pigtail catheter, the sheath is advanced to
the ostium where a stable position is obtained and documented
IMPLANTATION by angiography.
Typically, three vascular access sheaths are placed; these The current sheath has marker bands at its distal tip that
include two femoral venous sheaths, one for the appendage facilitate placement of the device. The device is selected on the
occlusion device itself, one for the intracardiac ultrasound basis of the dimension of the orifice determined by contrast
catheter, and one for arterial access. The arterial access sheath angiography as well as TEE, although as previously mentioned,
should be small (e.g., 4 Fr); this sheath allows placement of a CT image analysis prior to the procedure may be helpful in
pigtail catheter to identify the level of the aortic valve for selecting the optimal device size (Fig. 45.3). The device is
optimizing the trans-septal procedure as well as for monitoring advanced being careful to minimize the chance for entrapment
aortic pressure. The venous sheaths placed need to accommo- of air. Avoidance of trapping air is best accomplished by a
date the intracardiac ultrasound catheter using an 8- to 10-Fr high-pressure saline line that infuses saline during advance-
device as well as the occlusion device. The venous access for ment of the device. Once the device is positioned in the left
the occlusion device should be from the right femoral vein atrial appendage and confirmation has been obtained that there
because that facilitates trans-septal puncture; the site should be is no damage to the left atrial appendage, the sheath is with-
prepared so that eventually a 14-Fr sheath can be placed. The drawn leaving the occlusion device in place. If the device is
other two sheaths are usually placed on the contralateral left positioned too deeply in the left atrial appendage, it can be
femoral side. recaptured and withdrawn to a more ideal position. If the
[18/6/010/21:42:51] [577–583]
device was noninferior to treatment with warfarin with respect

to a composite efficacy endpoint of freedom from all stroke,
both ischemic and hemorrhagic, cardiovascular death, and
systemic embolization during follow-up out to 1500 patient-
years. Analysis of this trial continues. The results of this
randomized trial will be important to assess both safety and
efficacy of this device and approach. On the basis of the
outcome of that analysis, the number of patients that could
be treated if an approved device was available would increase
dramatically.
Fountain et al. (40) evaluated the frequency with which
the Watchman device might be used if the randomized
trial resulted in FDA approval. They used a screening log of
1798 patients to assess this. Of this group of patients who were
screened, only 31 patients were enrolled in the randomized
Figure 45.6 Echocardiography at implantation documenting that trial. Among the patients who were not enrolled in the random-
the device is positioned suboptimally and protrudes into the LA. ized trial, 21% would have been excluded from device use even
Recapture and deployment of another device is indicated. Abbre- after potential approval. The most common reason was the
viation: LA, left atrium.
requirement for long-term anticoagulation either because of a
prosthetic cardiac valve or for some other reason. The remain-
ing patients (79%) would have been candidates for the device.
Given the large number of patients with atrial fibrillation, the
device is positioned so that it protrudes too far into the left number of patients who could be candidates for the device such
atrium, then the device needs to be completely removed and a as this would be substantial.
new device placed (Fig. 45.6). Identifying the orientation of the
device with the ostium of the left atrial appendage is best
achieved using echocardiography. If a satisfactory axial posi- PLAATO
tion has been secured, angiography is performed to document The PLAATO (38,45) occluder (Fig. 45.7) was the first device
that there has been compression of the device, which is with a designed for percutaneous left atrial appendage occlusion. This
goal of compression of the dimensions of the unconstrained device was a self-expandable nitinol cage that was covered
device by 10% to 20%. Angiography through the sheath is used with an impermeable e-PTFE membrane. Similar to the Watch-
to assess the degree of obliteration of left atrial appendage flow. man device, it had anchors to prevent embolization and came
Finally, then a “tug test” is performed during which the oper- in different sizes ranging from 20 to 32 mm to ensure optimal
ator places manual traction on the device to ensure that it is sealing. This device was also delivered by a trans-septal
stable. During this time, TEE can document that the left atrial approach as described above using a special 12-Fr sheath.
appendage wall invaginates during device traction. Following Although this device is no longer marketed because of com-
this, the device is released and the sheath withdrawn. mercial issues, there are important data available regarding the
Following the procedure, with the current protocol, patient population in which it was used and its outcome.
patients are treated with aspirin and warfarin for 45 days Ostermayer et al. (45) reported on the outcome of two interna-
until a scheduled echocardiographic follow-up. If that follow- tional multicenter feasibility trials that included 111 patients.
up echocardiogram documents successful exclusion of the left All of these patients had a contraindication for anticoagulant
atrial appendage, the regiment is changed to aspirin and therapy and were at risk for stroke. Implantation of the
clopidogrel. After six months, the clopidogrel can be discon- PLAATO device was successful in 108 patients. Three patients
tinued if the clinical condition is satisfactory. required pericardiocentesis for hemopericardium and one
The initial worldwide experience with this specific device
was assessed in 75 patients who met all of the inclusion and
exclusion criteria for the trial (43). Of these 75 patients, a
Watchman device was implanted in 66 (88%). The most com-
mon reason for failure to implant the device was the fact that
left atrial appendage anatomy was not suitable. In the full
group of 75 patients, pericardial effusions were documented
in 2 (2.6%)—this was typically related to the trans-septal pro-
cedure. In this group, two devices embolized and were
retrieved percutaneously. Mean follow-up was available for
740 341 days. During this time, 91.7% of patients were able to
discontinue warfarin. There were no ischemic strokes or sys-
temic embolizations. Two patients had a limited transient
ischemic attack.
A multicenter randomized trial with this device has been Figure 45.7 The PLAATO device (no longer manufactured) is a
completed (including 707 patients) (40). This trial randomized self-expanding nitinol cage that in contrast to the Watchman device
patients with atrial fibrillation and a CHADS-2 (41) score 1 in has an occlusive polytetrafluoroethylene membrane. It also has
a 2:1 design to either the Watchman device or warfarin control. small anchors to prevent embolization.
The primary endpoint of this trial was to determine if the
[18/6/010/21:42:51] [577–583]
patient required urgent cardiovascular surgery and subse- device embolization immediately after implantation was iden-
quently died of neurologic causes. The primary endpoint for tified in four, and two patients had to undergo surgery for this.
these registries was the incidence of major adverse events Another occluder embolized during follow-up and was snared
defined as a composite of stroke, death, myocardial infarction, using percutaneous techniques. At that time of that presenta-
or need for procedural related cardiovascular surgery during tion, no patient had experienced a stroke during a follow-up
follow-up. All of these patients received aspirin and clopidog- time of 30 patient years. The Amplatzer device is currently
rel initially but no warfarin. Long-term treatment varied but at being modified for optimizing placement in the left atrial
a minimum consisted of aspirin-administered indefinitely. Dur- appendage and will be studied in a subsequent randomized
ing follow-up that averaged 9.8 months, there were six deaths clinical trial.
(5.4%) and the observed annual stroke rate was 2.2%. On the
basis of the baseline CHADS-2 score, this 2.2% event rate was
improved compared to the predicted 6.3% rate. OPEN SURGICAL LIGATION
In a larger experience of 210 patients treated with the Open surgical ligation has also been studied and is the focus of
PLAATO device worldwide (36) device implantation success a multicenter randomized trial. The approach that has been
was also high at 97.6%. In this group during a follow-up of 258 documented uses either a suture or a stapling device for closure
patient years, only 5 patients had a stroke that was again less at the time of elective coronary artery bypass graft surgery. A
than predicted by the baseline CHADS-2 score. pilot study (Left Atrial Appendage Occlusion Study, LAAOS)
has been reported (46). In this study, 97 patients were con-
sented but 20 had unsuitable left atrial appendage anatomy.
AMPLATZER OCCLUSION DEVICES This was predominantly because the left atrial appendage was
Amplatzer atrial septal occluder devices have also been used either too broad or too small. In the 77 patients who were
(Fig. 45.8) (37,42). These double disk devices are made of a randomized, 52 had occlusion of the left atrial appendage and
nitinol wire frame mesh with Dacron patches inside. Using a 25 patients served as controls. During surgery, complications
standard trans-septal approach, an introducer sheath compat- occurred with a torn left atrial appendage in nine patients. Left
ible with the chosen occluder device is advanced deeply within atrial appendage occlusion was achieved in only 45% of
the left atrial appendage. Typically, the left side of the device is patients in which sutures were used and 72% of patients
deployed within the left atrial appendage itself, while the right where a stapler approach was used. For a mean follow-up of
disk is deployed at the left atrial side of the ostium. The 13 7 months, 2.6% of patients had thromboembolic events.
position is ascertained by hand injection of contrast medium This pilot study has been used to plan a larger ongoing study.
through the sheath. Intracardiac ultrasound can be used to also
confirm device position. If the position is acceptable, the device
is released. In an initial experience of 16 patients (42), there was EXTRACARDIAC OBLITERATION
one technical failure with device embolization that required An additional invasive approach includes thorascopic extrac-
surgery. Subsequent to this, a somewhat larger experience of ardiac obliteration. An initial experience with 15 patients was
27 patients has been reported. In this series of 27 patients, reported by Blackshear et al. (47). With this approach, a double
lumen intratracheal tube is used. Selective intubation and ven-
tilation of the right lung results in reduction in the volume of
the left lung. Using video-assisted thorascopic instruments,
the pericardium is opened and the tip of the left atrial append-
age can be grasped. A loop is then manipulated to position it at
the base of the appendage where it is cinched to occlude the
appendage. In this small series, success was achieved in 14 of
15 patients; one patient developed bleeding and was converted
to open thoracotomy. In all of these patients, there was a
contraindication to anticoagulant therapy.
Patients were followed up for a mean of 42 14 months.
One fatal stroke occurred 55 months after surgery and one
nondisabling stroke 3 months after surgery. There were two
other deaths. In a subgroup of 11 patients with a history of
prior thromboembolism, there was an annualized rate of stroke
of 5.2% per year, which compared with a historical control rate
of 13% per year for similar aspirin-treated patients from the
Stroke Prevention in Atrial Fibrillation trials.
SUBXIPHOID TRANSPERICARIAL APPROACH

A final approach that is being developed is a subxiphoid
Figure 45.8 Amplatzer septal occlusion devices (AGA Medical transpericardial approach. Using this, the pericardial space is
Corporation, Plymouth, Minnesota, U.S.) have also been used “off entered percutaneously. A videoscope is inserted. A grasping
label” to occlude the left atrial appendage. Typically, the distal disk is device is manipulated around the pericardial space to the
deployed within the left atrial appendage (LAA) and the proximal region of the left atrial appendage and is used to capture the
disk is deployed just at the ostium. tip of that structure. Following this, similar to the thorascopic
approach described earlier, a loop is positioned at the base of
[18/6/010/21:42:51] [577–583]
the left atrial appendage and tightened to occlude that struc- Clopidogrel Trial with Irbesartan for prevention of Vascular
ture. There is limited data available using this approach. Events (ACTIVE W): a randomised controlled trial. Lancet 2006;
367:1903–1912.
17. Connolly SJ, Laupacis A, Gent M, et al. Canadian Atrial Fibrilla-
CONCLUSIONS tion Anticoagulation (CAFA) Study. J Am Coll Cardiol 1991; 18:
Stroke is one of the most feared complications of cardiovascular 349–355.
18. Ezekowitz MD, Bridgers SL, James KE, et al. Warfarin in the
disease because of its attendant morbidity and mortality. Pre-
prevention of stroke associated with nonrheumatic atrial fibrilla-
vention of stroke has received great attention. The relationship tion. Veterans Affairs Stroke Prevention in Nonrheumatic Atrial
between stroke, atrial fibrillation, and increasing age has been Fibrillation Investigators. N Engl J Med 1992; 327:1406–1412.
clearly identified. Multiple series have documented that in the 19. Fang MC, Go AS, Hylek EM, et al. Age and the risk of warfarin-
setting of nonvalvular atrial fibrillation that thrombus develops associated hemorrhage: the anticoagulation and risk factors in
and originates in the left atrial appendage. Given the constel- atrial fibrillation study. J Am Geriatr Soc 2006; 54:1231–1236.
lation of stroke, atrial fibrillation and left atrial appendage, 20. Gage BF, Boechler M, Doggette AL, et al. Adverse outcomes and
multiple efforts are currently aimed as exclusion of the latter. If predictors of underuse of antithrombotic therapy in medicare
successful left atrial appendage exclusion is documented to be beneficiaries with chronic atrial fibrillation. Stroke 2000; 31:
effective in preventing subsequent strokes and safe, then this 822–827.
21. Gage BF, van Walraven C, Pearce L, et al. Selecting patients with
approach will be widely used in patients who do not want to
atrial fibrillation for anticoagulation: stroke risk stratification in
take coumadin indefinitely or in patients who cannot take patients taking aspirin. Circulation 2004; 110:2287–2292.
coumadin. 22. Go AS, Hylek EM, Borowsky LH, et al. Warfarin use among
ambulatory patients with nonvalvular atrial fibrillation: the anti-
coagulation and risk factors in atrial fibrillation (ATRIA) study.
REFERENCES Ann Intern Med 1999; 131:927–934.
1. Rosamond W, Flegal K, Furie K, et al. Heart disease and stroke 23. Go AS, Hylek EM, Chang Y, et al. Anticoagulation therapy for
statistics—2008 update: a report from the American Heart Asso- stroke prevention in atrial fibrillation: how well do randomized
ciation Statistics Committee and Stroke Statistics Subcommittee. trials translate into clinical practice? JAMA 2003; 290:2685–2692.
Circulation 2008; 117:e25–e146. 24. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic
2. Benjamin EJ, Levy D, Vaziri SM, et al. Independent risk factors for therapy to prevent stroke in patients who have nonvalvular atrial
atrial fibrillation in a population-based cohort. The Framingham fibrillation. Ann Intern Med 2007; 146:857–867.
Heart Study. JAMA 1994; 271:840–844. 25. Hylek EM, Evans-Molina C, Shea C, et al. Major hemorrhage and
3. Garcia DA, Hylek E. Reducing the risk for stroke in patients who tolerability of warfarin in the first year of therapy among elderly
have atrial fibrillation. Cardiol Clin 2008; 26:267–275, vii. patients with atrial fibrillation. Circulation 2007; 115:2689–2696.
4. Lin HJ, Wolf PA, Kelly-Hayes M, et al. Stroke severity in atrial 26. Petersen P, Boysen G, Godtfredsen J, et al. Placebo-controlled,
fibrillation. The Framingham Study. Stroke 1996; 27:1760–1764. randomised trial of warfarin and aspirin for prevention of throm-
5. Lloyd-Jones DM, Wang TJ, Leip EP, et al. Lifetime risk for devel- boembolic complications in chronic atrial fibrillation. The Copen-
opment of atrial fibrillation: the Framingham Heart Study. Circu- hagen AFASAK study. Lancet 1989; 1:175–179.
lation 2004; 110:1042–1046. 27. Waldo AL, Becker RC, Tapson VF, et al. Hospitalized patients
6. Miyasaka Y, Barnes ME, Gersh BJ, et al. Secular trends in inci- with atrial fibrillation and a high risk of stroke are not being
dence of atrial fibrillation in Olmsted County, Minnesota, 1980 to provided with adequate anticoagulation. J Am Coll Cardiol 2005; 46:
2000, and implications on the projections for future prevalence. 1729–1736.
Circulation 2006; 114:119–125. 28. Albers GW, Diener HC, Frison L, et al. Ximelagatran vs. warfarin
7. Onalan O, Crystal E. Left atrial appendage exclusion for stroke for stroke prevention in patients with nonvalvular atrial fibrilla-
prevention in patients with nonrheumatic atrial fibrillation. Stroke tion: a randomized trial. JAMA 2005; 293:690–698.
2007; 38:624–630. 29. Olsson SB. Stroke prevention with the oral direct thrombin inhib-
8. Psaty BM, Manolio TA, Kuller LH, et al. Incidence of and risk factors itor ximelagatran compared with warfarin in patients with non-
for atrial fibrillation in older adults. Circulation 1997; 96:2455–2461. valvular atrial fibrillation (SPORTIF III): randomised controlled
9. Ruigomez A, Johansson S, Wallander MA, et al. Incidence of trial. Lancet 2003; 362:1691–1698.
chronic atrial fibrillation in general practice and its treatment 30. Blackshear JL, Odell JA. Appendage obliteration to reduce stroke
pattern. J Clin Epidemiol 2002; 55:358–363. in cardiac surgical patients with atrial fibrillation. Ann Thorac
10. Stewart S, Hart CL, Hole DJ, et al. Population prevalence, inci- Surg 1996; 61:755–759.
dence, and predictors of atrial fibrillation in the Renfrew/Paisley 31. Budge LP, Shaffer KM, Moorman JR, et al. Analysis of in vivo left
study. Heart 2001; 86:516–521. atrial appendage morphology in patients with atrial fibrillation: a
11. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an inde- direct comparison of transesophageal echocardiography, planar
pendent risk factor for stroke: the Framingham Study. Stroke 1991; cardiac CT, and segmented three-dimensional cardiac CT. J Interv
22:983–988. Card Electrophysiol 2008; 23:87–93.
12. The effect of low-dose warfarin on the risk of stroke in patients 32. Ernst G, Stollberger C, Abzieher F, et al. Morphology of the left
with nonrheumatic atrial fibrillation. The Boston Area Anticoagu- atrial appendage. Anat Rec 1995; 242:553–561.
lation Trial for Atrial Fibrillation Investigators. N Engl J Med 1990; 33. Veinot JP, Harrity PJ, Gentile F, et al. Anatomy of the normal left
323:1505–1511. atrial appendage: a quantitative study of age-related changes in
13. Stroke Prevention in Atrial Fibrillation Study. Final results. Cir- 500 autopsy hearts: implications for echocardiographic examina-
culation 1991; 84:527–539. tion. Circulation 1997; 96:3112–3115.
14. Optimal oral anticoagulant therapy in patients with nonrheumatic 34. Kerut EK. Anatomy of the left atrial appendage. Echocardiogra-
atrial fibrillation and recent cerebral ischemia. The European phy 2008; 25:669–673.
Atrial Fibrillation Trial Study Group. N Engl J Med 1995; 333:5–10. 35. Meissner I, Whisnant JP, Khandheria BK, et al. Prevalence of
15. Bungard TJ, Ghali WA, Teo KK, et al. Why do patients with atrial potential risk factors for stroke assessed by transesophageal
fibrillation not receive warfarin? Arch Intern Med 2000; 160:41–46. echocardiography and carotid ultrasonography: the SPARC
16. Connolly S, Pogue J, Hart R, et al. Clopidogrel plus aspirin versus study. Stroke Prevention: Assessment of Risk in a Community.
oral anticoagulation for atrial fibrillation in the Atrial fibrillation Mayo Clin Proc 1999; 74:862–869.
[18/6/010/21:42:51] [577–583]
36. Bayard YL, Ostermayer SH, Hein R, et al. Percutaneous devices 43. Sick PB, Schuler G, Hauptmann KE, et al. Initial worldwide
for stroke prevention. Cardiovasc Revasc Med 2007; 8:216–225. experience with the WATCHMAN left atrial appendage system
37. Cruz-Gonzalez I, Cubeddu RJ, Sanchez-Ledesma M, et al. Left for stroke prevention in atrial fibrillation. J Am Coll Cardiol 2007; 49:
atrial appendage exclusion using an Amplatzer device. Int J 1490–1495.
Cardiol 2009; 134:e1–e3. 44. Sievert H, Lesh MD, Trepels T, et al. Percutaneous left atrial
38. El-Chami MF, Grow P, Eilen D, et al. Clinical outcomes three years appendage transcatheter occlusion to prevent stroke in high-risk
after PLAATO implantation. Catheter Cardiovasc Interv 2007; 69: patients with atrial fibrillation: early clinical experience. Circula-
704–707. tion 2002; 105:1887–1889.
39. Fountain R, Holmes DR Jr., Hodgson PK, et al. Potential applica- 45. Ostermayer SH, Reisman M, Kramer PH, et al. Percutaneous left
bility and utilization of left atrial appendage occlusion devices in atrial appendage transcatheter occlusion (PLAATO system) to
patients with atrial fibrillation. Am Heart J 2006; 152:720–723. prevent stroke in high-risk patients with non-rheumatic atrial
40. Fountain RB, Holmes DR, Chandrasekaran K, et al. The PROTECT fibrillation: results from the international multi-center feasibility
AF (WATCHMAN Left Atrial Appendage System for Embolic trials. J Am Coll Cardiol 2005; 46:9–14.
PROTECTion in Patients with Atrial Fibrillation) trial. Am Heart J 46. Healey JS, Crystal E, Lamy A, et al. Left Atrial Appendage
2006; 151:956–961. Occlusion Study (LAAOS): results of a randomized controlled
41. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical pilot study of left atrial appendage occlusion during coronary
classification schemes for predicting stroke: results from the bypass surgery in patients at risk for stroke. Am Heart J 2005; 150:
National Registry of Atrial Fibrillation. JAMA 2001; 285:2864–2870. 288–293.
42. Meier B, Palacios I, Windecker S, et al. Transcatheter left atrial 47. Blackshear JL, Johnson WD, Odell JA, et al. Thoracoscopic
appendage occlusion with Amplatzer devices to obviate antico- extracardiac obliteration of the left atrial appendage for stroke
agulation in patients with atrial fibrillation. Catheter Cardiovasc risk reduction in atrial fibrillation. J Am Coll Cardiol 2003; 42:
Interv 2003; 60:417–422. 1249–1252.
[2/7/010/10:12:57] [584–592]
46
Percutaneous closure of atrial septal defect

and patent foramen ovale
Yves Laurent Bayard, Andreas Wahl, and Bernhard Meier
INTRODUCTION the lung vessels. The possibility of a transient right-to-left shunt

Over the past decades, transcatheter techniques for closure of during Valsalva maneuvers includes the risk of paradoxical
atrial septal defects (ASD) and patent foramen ovale (PFO) have embolism via the ASD.
emerged not only as an alternative to surgery, but as therapy of
first choice in most patients. While the first interventional devices
for this purpose were afflicted with rather high rates of compli-
Indications for ASD Closure
Regardless whether the patient is symptomatic, defect closure is
cations and residual shunts, current occlusion systems achieve
indicated if a significant left-to-right shunt (Qp/Qs ratio > 1.5:1)
high occlusion rates and complications are rare.
is measured during right heart catheterization or a right heart
dilatation (right ventricle end-diastolic diameter >30 mm) can be
ATRIAL SEPTAL DEFECT detected in transthoracic echocardiography (TTE). In addition,
Anatomical Considerations ASDs may be closed to prevent paradoxical embolism, even if
ASD within the oval fossa (ASD of secundum type) occur in 1.5 there is no evidence for hemodynamic relevance. As small ASDs
to 2 of 1000 newborns and are two to three times more common are easy to close percutaneously, the indication for closure can be
in females (1,2). They are the most common ASDs (about 70%), generalized.
and may be singular or multiple. Implantation of a transcath-
eter device requires sufficient cranial, caudal, and posterior
rims. However, an anterior rim (behind the aortic root) is not PATENT FORAMEN OVALE
necessary. Furthermore, a minimal distance of the ASD mar- Anatomical Considerations
gins to neighbor structures is required (right upper pulmonary The interatrial septum is formed by two structures, the left-
vein, coronary sinus, atrioventricular valves). To determine sided more fibrous septum primum, and the right-sided mus-
whether these morphologic conditions for interventional ASD cular septum secundum. After having grown from the periph-
closure are given, a transesophageal echocardiogram with ery to the center, they form a slit valve that opens with pressure
measurement of the atrial dimensions is required prior to from the right. When pulmonary circulation is established after
closure in most adults. Not suitable for catheter treatment are birth and left atrial pressure exceeds right atrial pressure, the
ostium primum defects or sinus venosus defects. valve shuts. In most individuals, the overlapping parts of the
septa fuse permanently. Nevertheless, the prevalence of PFO is
high. Autopsy studies have shown a prevalence of up to 30% in
Pathophysiology healthy subjects (3). In these individuals, the PFO permits
Defects of the atrial septum represent the second most common intracardiac shunting during periods when right atrial pressure
congenital heart disease after ventricular septal defects. Many exceeds left atrial pressure.
patients with an isolated secundum ASD do not develop any In 50% to 85% of subjects with an atrial septal aneurysm
clinical symptoms during childhood. Therefore, the diagnosis (ASA), a PFO is present. This congenital abnormality of the
of ASD is frequently missed until adult life. Patients may atrial septum is formed by an elongated, amuscular membrane
present with unspecific symptoms such as dyspnea, fatigue, in the region of the fossa ovalis. ASA is usually defined as
or palpitations. Typical auscultation findings are a soft ejection- excursion of the atrial septum 10 mm in either direction or
type systolic murmur resulting from the relative pulmonary both directions combined. Additionally, the base of the flimsy
stenosis and a characteristic wide and fixed splitting of the portion of the interatrial septum must exceed 15 mm (4).
second heart sound. In secundum type ASD, the electrocardio-
gram may show right-axis deviation and incomplete right
bundle branch block, whereas left-axis deviation might be Pathophysiology
noted in ostium primum defects. Chest X-ray may reveal In contrast to an ASD, a PFO alone is not considered a
prominent pulmonary arteries and right heart enlargement. pathologic finding. A possible relationship between PFO and
Endocarditis is not an issue because the shunt between low stroke was first suggested by Cohnheim in 1877 (5). Nowa-
pressure chambers is free of turbulence. days, PFO and ASA when associated with PFO have been
Left-to-right shunt patients with an ASD are at-risk for recognized as potential mediators of several disease manifes-
right heart overload with consecutive cardiac arrhythmias or tations, including paradoxical embolism, orthostatic desatura-
heart failure. In addition, irreversible injury of the pulmonary tion in the setting of the platypnea-orthodeoxia syndrome,
vascular bed with secondary pulmonary hypertension or even refractory hypoxemia due to right-to-left shunt in patients
Eisenmenger’s syndrome can result from volume overload of with right ventricular infarction or severe pulmonary disease,
[2/7/010/10:12:57] [584–592]
PERCUTANEOUS CLOSURE OF ATRIAL SEPTAL DEFECT AND PATENT FORAMEN OVALE 585
neurologic decompression illness in divers, high-altitude of a multiperforated septum. In case of a PFO, TEE in combi-
pulmonary edema (6), and migraine with aura. nation with a Valsalva maneuver is the most sensitive and
In 35% to 40% of all patients with an ischemic stroke the specific instrument to prove right-to-left shunting through the
cause of the stroke remains “cryptogenic” (7). Several studies structure, to recognize the presence of ASA, and to estimate the
have shown that the prevalence of a PFO in these patients is severity of the shunt as additional risk factors for paradoxical
significantly higher than in control groups. The incidence of a embolism.
PFO in patients with cryptogenic stroke varies between 44% During the intervention, patients are administered hep-
and 66% compared with only 9% to 27% in the control groups arin and endocarditis prophylaxis (e.g., cefuroxim IV). Echo-
(8–11). Therefore, the PFO should be considered as the source of cardiographic guidance is not recommended for PFO closure. It
the embolic event, even if the thrombus crossing the atrial is not required for a safe procedure, but prolongs the procedure
septum can only be shown very rarely (12,13). Serena and considerably and increases complications (18). For ASD closure
Davalos found that large shunts in PFO patients are associated the use of either TEE or intracardiac echocardiography (ICE)
with a higher risk for stroke than small shunts (7). guidance is debatable. When TEE guidance is used, atropine to
ASA in the absence of PFO is not a risk factor for strokes prevent hypersalivation is given. Conscious sedation (midazo-
(14). However, ASA is an important reason for the foramen not lam, propofol) or general anesthesia with endotracheal intuba-
closing and has been suggested to increase the diameter of a tion is used.
present PFO because of the highly mobile atrial septal tissue. Venous access is gained via the right femoral vein. The
This leads to a more frequent and wider opening of the channel shunt is passed directly with the guidewire or with a multi-
(15). ASA or a Eustachian valve might as well promote a right- purpose catheter. Balloon sizing (Fig. 46.1) with a soft measuring
to-left shunt by directing the blood flow from the inferior vena balloon is recommended for ASDs. The waist of the balloon
cava toward the PFO (16). Mas et al. found a significantly caused by the margins of the ASD is measured to determine the
higher risk for transient ischemia attack (TIA) or stroke in size of the defect (stretched diameter) and choose the appropriate
patients with a PFO in combination with an ASA of about device size. If the balloon sits stable in the septum, the device is
20% at four years compared with patients with PFO only (17). oversized by 20% to 40%, if not by 40% to 60%, assuming a
nonrobust rim.
The further implantation technique of the respective
Indications for PFO Closure occluders is described in the corresponding chapters later.
It is not scientifically supported, but common, to exclude all
Generally, for larger ASDs only the Amplatzer ASD Occluder
other potential sources for the embolic event such as atrial
should be used, for small ASDs or PFOs other systems are also
fibrillation, carotid stenosis, or thrombophilia before closing a
suitable. Only the Amplatzer Septal Occluder comes in an ASD
PFO. Transcatheter PFO closure is still considered to be indi-
and a PFO version.
cated only after an otherwise unexplained (cryptogenic)
If the procedure is conducted under echocardiographic
embolic stroke, TIA, or peripheral embolism, if the potential
guidance, a contrast study with Valsalva maneuver is per-
for right-to-left shunt via a PFO can be demonstrated by con-
formed after device implantation to evaluate the immediate
trast transesophageal echocardiography (TEE).
closure rate. If not, a right atrial dye injection confirms position
and informs about the residual shunt (Fig. 46.2). Hemostasis is
Preparations, Implantation Technique, achieved by manual compression of the femoral vein. The
and Follow-up Regimen procedure can be performed on an outpatient basis and may
If an atrial septal communication is suspected, TEE is recom- take <30 minutes, with a fluoroscopy time of about five
mended to anatomically define the defect, measure its dimen- minutes. The patient can resume unrestricted physical activity
sions and margins, for quantification of shunting and exclusion a few hours post procedure.
Figure 46.1 Balloon sizing of an ASD. The waist of the

balloon caused by the margins of the ASD (30 mm) is
used to choose the appropriate device size (36 mm). The
inset shows the perfect closure at a bubble test during a
follow-up transesophageal echocardiogram at six months.
Abbreviations: ASD, atrial septal defect; LA, left atrium;
RA, right atrium; LV, left ventricle.
[2/7/010/10:12:57] [584–592]
prevails. Intratunnel devices, percutaneous suture devices, or

radio frequency fusion devices are in clinical evaluation. They
are hampered by nonapplicability in the most ominous PFOs,
the ones with little overlap and an associated ASA.
Amplatzer Occluder
The Amplatzer Septal Occluder (AGA Medical Corporation,
Golden Valley, Minnesota, U.S.) is the most frequently
implanted ASD device (Fig. 46.3). The device consists of a
double-disk system and is available in diameters up to 40 mm.
It is manufactured from a wire frame mesh with Dacron
patches inside. The two disks are linked together by a short,
elastic connecting waist. In the ASD version, the diameter of the
waist corresponds to the size of the ASD to allow “stenting” of
the defect and self-centering (important also during implanta-
tion). This unique feature allows closure of even large defects
(up to 40 mm stretched diameter), which cannot be closed by
other devices. On the one hand, the rather rigid wire frame
includes the advantage of excellent stability even in defects
without anterior septal rim. On the other hand, this feature
involves a certain risk of injury to cardiac structures such as the
free atrial wall pounding against the aortic root. Some authors
expressed their concerns about the device’s content of nickel.
Figure 46.2 Angiographic assessment before release of a 25-mm However, in 2002, Kong published a study investigating the
Amplatzer PFO Occluder (AGA Medical Corporation, Golden Valley, chemical stability of the implant (25). He found no measurable
Minnesota, U.S.). A projection depicting the device in perfect profile is elevation of nickel in the blood either in vitro or in vivo (animal
selected. The salient structures are easy to identify. The device resem- model, humans).
bles the arcade figure Pacman biting into a dot which created the term The Amplatzer PFO Occluder (AGA Medical Corporation)
“Pacman” sign (18). Abbreviations: PFO, patent foramen ovale; LA, left
is the most commonly used device for PFO closure (Fig. 46.3).
atrium; RA, right atrium; SP, septum primum; SS, septum secundum.
It is available in various sizes: 18, 25, 30, 35, and 40 mm. In
most PFO devices, the right atrial disk is larger than the left
atrial disk. The connecting waist does not need to stent the
Acetylsalicylic acid 100 mg and clopidogrel 75 mg are
defect and is therefore thin. However, it needs to be stretch-
prescribed for a minimum of five and one months after implan-
able for long tunnels. The so-called cribriform devices feature
tation, respectively. Endocarditis prophylaxis is recommended
two identical disks with a thin waist. They are suited for PFOs
for two to six months. Follow-up is performed with TTE to
or multifenestrated ASDs.
confirm device position before hospital discharge and with TEE
A long curved sheath (5–13 Fr, depending on the size of
after six months to confirm complete closure of the intra-atrial
the device) is advanced through the ASD or PFO into the left
communication and exclude thrombosis of the device (19). If
atrium. According to the stretched diameter of an ASD or
the PFO proves completely closed, no further follow-up mea-
echocardiographic features of a PFO, a suitable device size is
sures need to be planned. In case of persistence of a relevant
chosen. Thereafter, the collapsed device attached to a specific
residual shunt after six months, later spontaneous closure is
delivery catheter is advanced through the long sheath up to the
unlikely. With the exception of tiny transdevice residual
tip. By pulling back the long sheath while keeping the delivery
shunts, the risk for paradoxical embolisms persists and so
catheter in position, the left atrial part of the device is allowed
does the indication for blood thinning. Implantation of a second
to self-expand. The expanded left atrial disk is pulled back
device is feasible, resulting in complete closure in most cases.
toward the atrial septum together with the sheath as a unit. For
ASDs the connecting waist is expanded before approaching
Double-Disk Devices the septum to assure self-centering. By further pulling back the
In 1974, King and Mills (20) used double-umbrella devices to long sheath the right atrial disk is released. After checking the
occlude experimentally created ASDs in a canine model. In implant’s stability by wiggling the pusher wire and control of
1976, they successfully performed this technique in a 17-year- the correct position by angiography (Fig. 46.2) and echocar-
old female (21). After having treated five additional patients diography, the device is released from the pusher wire.
they reported excellent long-term results (22) in 1984 as well as Transcatheter closure using the Amplatzer devices has
in 2003 (23). Transcatheter PFO closure was first described by shown promising results. Berger et al. reported their experience
Bridges et al. in 1992 (24). They used the Clamshell device in with the Amplatzer Septal Occluder in 200 patients (26); 68 of
36 patients after presumed paradoxical embolism. Since then, them had a PFO. They claimed complete closure of the defect in
the usage of transcatheter closure devices has became more and all patients without adverse events. More recently, Sievert and
more popular and is now a routine procedure in many centers colleagues reported a larger series with 650 patients who
all over the world. received an Amplatzer Septal Occluder between 1997 and
A number of occlusion systems have been developed. 2005 (27). Implantation success rate was 98%. During follow-
Although current occluder design has evolved since the first up of up to eight years, complete defect closure could be
devices, the principle of double-umbrella devices with one disk documented in 96% of the patients with a single ASD and
each on the right and the left atrial side of the defect still 71% of the patients with a multiperforated septum. There were
[2/7/010/10:12:57] [584–592]
Figure 46.3 Occluder overview. (A and B) Amplatzer ASD Occluder (AGA Medical Corporation, Golden Valley, Minnesota, U.S.); (C and D)
Amplatzer PFO Occluder (AGA Medical Corporation); (E) STARFlex Occluder (NMT Medical, Inc., Boston, Massachusetts, U.S.);
(F) BioSTAR Occluder (NMT Medical, Inc.); (G) Helex Septal Occluder (W. L. Gore & Associates, Inc., Flagstaff, Arizona, U.S.);
(H) Atriasept Occluder (Cardia, Inc., Eagan, Minnesota, U.S.); (I) Occlutech Occluder (Occlutech, Jena, Thuringia, Germany); (K) Solysafe
Occluder (Swissimplant AG, Solothurn, Switzerland); (L) Premere Occluder (St. Jude Medical, Inc., St. Paul, Minnesota, U.S.); (M) SeptRX
Occluder (Secant Medical, Perkasie, Pennsylvania, U.S.).
[2/7/010/10:12:57] [584–592]
two device embolizations but the devices could be recovered Helex Occluder
without need for surgery. The most common complication The Helex Occluder (W. L. Gore & Associates, Inc., Flagstaff,
during follow-up was new onset of atrial fibrillation in 4%. Arizona, U.S.) is a device with a unique spiral shape. It consists
We reported a series of 620 patients who underwent PFO of two disks connected in the center by an eyelet mechanism
closure using the Amplatzer PFO Occluder (18). All implanta- (Fig. 46.3). It is available in diameters of up to 35 mm. Both
tions were successfully conducted without echocardiographic disks are formed by one continuous wire (nitinol) in the shape
guidance. Procedural complications occurred only in five of a spiral. An expanded polytetrafluoroethylene patch is
patients: one patient sustained a TIA and four showed arterio- attached to this wire. Bringing the wire in a straight configu-
venous fistulae at the puncture site, requiring elective surgical ration allows loading inside a 9-Fr delivery sheath. The spiral
correction. Mean follow-up time was three years. Freedom design provides a very low profile and an atraumatic healing
from recurrent ischemic stroke, TIA, or peripheral embolism process after implantation. As another unique feature, the
was 99% at one year, 99% at two years, and 97% at five years Helex implantation system has a security cord attached to the
post procedure. device which allows retrieval at any point during the proce-
dure, even after initial release from the delivery catheter.
CardioSEAL/STARFlex/BioSTAR Before implantation the Helex device attached to a con-
CardioSEAL (NMT Medical, Inc., Boston, Massachusetts, U.S.) trol catheter is loaded into a delivery catheter. The delivery
is a revised version of the Clamshell occluder. It can be catheter is advanced through the ASD until the tip containing
delivered via an 11-Fr sheath and is available in sizes up to the collapsed device is positioned inside the left atrium. The
38 mm. Two rectangular disks, each consisting of four wire device is advanced through the delivery catheter allowing the
spring arms, form the frame of this double-umbrella system. left atrial disk to expand in the left atrium. The sheath and
Each disk is covered with a knitted polyester patch. The centers the catheter are pulled back simultaneously until the released
of the umbrellas are connected. part is positioned close to the septum as seen in TEE. Thereafter,
STARFlex (NMT Medical, Inc.), a newer version of the the right atrial disk is opened. After control of a stable and
CardioSEAL, has an additional microspring system (Fig. 46.3): It correct position of the device by TEE, the device is detached.
also comes in a version with six arms with a diameter of 38 and Finally, the security cord is removed.
43 mm. The microsprings connect the distal tip of each arm with In 2007, Jones and colleagues reported the results of the
the opposing arm. This system allows some self-centering of the U.S. pivotal study on ASD closure with 119 Helex device patients
device inside an ASD. In addition, the micosprings provide a versus 128 surgical closure patients (30). In the interventional
close attachment of each disk to the septum, which allows for a group one device had to be retrieved after embolization. In the
low profile and is helpful in more complex defects (e.g., presence surgical group one patient died from cardiac tamponade. With
of ASA). The STARFlex device has the possibility to implant an implantation success rate of 92% and significant residual leak
relatively large devices into small defects. This is advantageous in in only 2% of the patients treated with the Helex device,
patients with a multiple perforated septum, just like the cribri- noninferiority to surgical closure could be demonstrated.
form Amplatzer occluders. The device can be introduced through In 2006, Billinger et al. reported their experience of PFO
a 10-Fr sheath, and the implantation technique resembles the closure using the Helex device in 128 patients (31). Device
procedure with the Amplatzer occluders to a large extent. implantation was successful in all but one patient. In a mean
Carminati et al. performed PFO closure using the follow-up time of 21 11 months, one device embolization and
CardioSEAL device in 79 patients and the STARFlex device in one TIA occurred. There were no strokes, deaths, perforations,
38 patients (28). Either device was successfully implanted in all or thrombus formations on the device. Other reports have
patients. A second device had to be placed immediately in six signaled relatively poor closure rates with this device (32).
patients. During follow-up, there were no complications and no
recurrent cerebral ischemia. STARFlex showed a significantly Atriasept Occluder
lower wire frame fracture rate than its predecessor, Cardio- The Atriasept Occluder (Cardia, Eagan, Minnesota, U.S.) is the
SEAL. Additionally, the occlusion rate was significantly higher newest generation of double-disk devices based on the PFO
using the newer STARFlex device. Star device from 1998 (Fig. 46.3). Atrisept is made of a nitinol
A larger series of PFO patients who received a STARFlex frame covered by two equally sized disks of polyvinyl alcohol
device has been reported by Taaffe and colleagues in 2008: 220 foam. The implantation technique resembles the Amplatzer
patients were implanted with a device; in two of them, the device device procedure. In 2008, Luermans and colleagues presented
had to be exchanged for another size. Post procedure, 10 patients the results of a European multicenter study including 430
developed atrial fibrillation, which resolved spontaneously in patients who had their PFO closed using the Atriasept Occluder
seven patients. Within 30 days after implantation, a thrombus (33). Device malpositioning occurred in 4 patients and new
formation was found on the occluder of eight patients. They were onset atrial arrhythmia was observed in 22 patients (5%). In
administered oral anticoagulation for three months, and all a median follow-up time of one year, two patients (0.5%)
thrombi resolved without clinical sequelae. sustained a stroke and seven patients (2%) a TIA.
The relatively high thrombosis rate has prompted the
manufacturer to replace the fabric with collagen. This new Occlutech Occluder
device is called BioSTAR because the collagen is bioabsorbable Occlutech is a meshwire nitinol occluder (Occlutech, Jena,
more rapidly and more completely than the fabric used pre- Thuringia, Germany) that copies the Amplatzer occluders
viously (Fig. 46.3). In the BioSTAR Evaluation Study (BEST), the both in appearance and implantation technique (Fig. 46.3).
device was successfully implanted in 57 of 58 patients (29). The main difference is the knitting technique of the Occlutech
There were no procedural complications. The occlusion rate device and the absence of a left atrial knob. Thus, the risk for
assessed by contrast TTE was 92% after one month and 96% thrombus formation on the device might be even lower than in
after six months. Amplatzer devices. In a first feasibility study, 37 PFO patients
[2/7/010/10:12:57] [584–592]
were enrolled. One patient could not be implanted because the 90 cm and is introduced via a 12-Fr sheath. The PFO is crossed
PFO could not be crossed with the catheter; one patient with the tip of the device. When drawing the system back, first
developed transient atrial fibrillation. Of the 34 patients seen the septum primum and then the septum secundum are caught
at six month follow-up, complete occlusion could be shown in and sewn together. Experience with this device in humans is
30 patients. There was one case of significant left-to-right shunt. very limited, however, a clinical trial is planned.
There was no recurrent stroke or peripheral embolism.
Solysafe Occluder BioTREK Device

The Solysafe Occluder (Swissimplant AG, Solothurn, Switzer- The BioTREK device (NMT Medical, Inc.) represents the next
land) consists of a device with two foldable polyester patches stage of development of the BioSTAR Occluder. Other than the
attached to eight metal wires, folding to a very flat-profiled BioSTAR with its nitinol frame, the BioTREK device is made of
double-disk occluder during intervention (Fig. 46.3). In a pilot poly-4-hydroxybutyrate, a completely bioresorbable material
study, the occluder was successfully implanted in 29 PFO and that is isolated from bacteria. The device is flexible, radiopaque,
15 ASD patients without procedural complications (34). Follow- and repositionable. Clinical studies are currently ongoing.
up TEE six months after implantation showed complete occlu-
sion of all interatrial communications. PFx Closure System
Despite the favorable safety and efficacy of permanent
Premere Occluder implant devices, apprehensions and disadvantages exist with
The Premere PFO Closure System (St. Jude Medical, Inc., St. a foreign structure that remains in the heart. Potential com-
Paul, Minnesota, U.S.) is a dual-anchor arm device (Fig. 46.3). plications associated with PFO implant devices include throm-
The anchors are made of nitinol and the right anchor is bus formation, device erosion, device embolization, and atrial
sandwiched between two layers of knitted polyester fabric. A fibrillation.
flexible polyester braided tether running through the center of The PFx Closure System (Cierra, Inc., Redwood City,
the anchor holds the two anchors together. The anchors are California, U.S.) allows device-free closure of PFOs, leaving
retrievable and repositionable. After delivery, they are locked no foreign material behind after the procedure is finished. It
together and the tether is cut. The distance between the two consists of a catheter with a metal electrode at the distal end
anchors is selectable before release, depending on the length of and an elastomeric distal housing covering the electrode. Elec-
the PFO track. The device comes in 15, 20, and 25 mm diameter. trical leads exit the proximal end of the device and connect to a
Buscheck and colleagues reported the results of the CLOSEUP radio frequency generator.
trial, a multicenter, nonrandomized study to determine the The PFx catheter is inserted through a 16-Fr sheath. Under
safety and effectiveness of the Premere device for PFO closure TEE guidance, the distal housing is moved against the atrial wall
(35). The procedure was technically successful in all 73 patients. so the electrode covers the PFO mouth. After confirmation of a
One patient developed transient atrial fibrillation. Six months correct device position, suction is applied. When adequate seal is
post procedure, follow-up TEE was conducted and showed achieved, the guidewire is removed and the radio frequency
complete PFO occlusion in 86% of the patients. There were no generator is engaged in accordance with a prescribed power
device dislocations, recurrent strokes, or deaths. algorithm resulting in heating of the cardiac tissue.
In 2009, Sievert and colleagues reported intermediate
results of PFO closure using the PFx system within a non-
NOVEL DEVICES AND NONDEVICES
randomized, multicenter study (36). In 130 of 144 patients
SeptRX Device and Coherex Flatstent
enrolled, radio frequency energy could be successfully applied.
The SeptRX device (Stout Medical Group, Perkasie, Pennsylva-
Besides a procedural bleeding complication with need for
nia, U.S.; Fig. 46.3) and the Coherex Flatstent (Coherex Medical,
transfusion, there were no acute or intermediate complications
Inc., Salt Lake City, Utah, U.S.) have been designed to plug the
of this technique such as recurrent strokes, deaths, conduction
tunnel of the PFO and expose as little foreign surface to either
abnormalities, or perforations following the procedure. In a
side of the atrial septum as possible. Thereby, the risk of
mean follow-up time of six months, complete occlusion of the
thrombus formation is minimized and the amount of foreign
PFO was achieved in only 55% of the patients. In patients with
material in the body is reduced.
a PFO stretched diameter of <8 mm, the occlusion rate was
The SeptRX device is made of a nitinol wire mesh and a
72%, suggesting that the vacuum created by the first generation
nitinol frame with small left and right atrial anchors. In an
of PFx devices to achieve tissue apposition might not be suffi-
initial feasibility study, complete occlusion of the PFO within
cient in larger PFOs. The device has since been pulled off the
the first 30 days after implantation could be achieved in only 6
market.
of 11 patients. After six months, complete occlusion could be
observed in all subjects.
The Coherex Flatstent device is designed to stent the PFO PERSPECTIVE
tunnel. So far, the self-expanding device is only suitable for A PFO may lead to paradoxical embolism and stroke. How to
tunnel lengths >4 mm and a balloon stretched diameter of 4 to treat patients with (cryptogenic) stroke and PFO is still discussed
10 mm. Clinical experience is limited to a few preliminary cases controversially. Possible treatments are antiaggregation or anti-
in humans. coagulation, surgery, or transcatheter closure of the defect. Med-
ical treatment has a risk of recurrent embolic events of 2% to 14%
Sutura Super Stitch Device per year (11,14,17,37–40). The annual recurrence for stroke and
The Sutura Super Stitch device is a suture-based closure device death is 1% to 7% (17,36,37,39). In addition, anticoagulation
(Sutura, Inc., Fountain Valley, California, U.S.) based on a punc- encounters a risk for hemorrhagic complications of 9% to 15%
ture site closure technique. The device has a working length of per year and 2% to 5% per year for severe bleeding (cerebral
[2/7/010/10:12:57] [584–592]
and/or need of a transfusion) (9,10). Khairy et al. have compared Overall, complete closure can be expected in about 90%
the medical and the interventional approach of stroke prevention of cases with Amplatzer devices and in 70% to 85% with other
in PFO patients (41) referring to multiple studies from 1985 to devices. The result at four to six months is final. So-called “late
2003. The one-year rate of recurrent neurologic thromboembo- closures” are usually a product of less sensitive screening
lism after transcatheter intervention was 0% to 5%, whereas techniques. The degree of residual shunt is conceptionally
medical management resulted in a one-year recurrence rate of likely to correlate with further paradoxical embolisms,
4% to 12%. However, comparability of these studies is limited although this was not universally confirmed by respective
because of general differences such as the incidence of risk analyses.
factors and baseline imbalances.
Surgical closure of a PFO can be performed at low risk
(8,42,43). Homma et al. reported their results in 28 patients (41). RANDOMIZED PFO CLOSURE STUDIES
Mean hospital stay was 8 4 days. 18% of these patients So far, no randomized PFO study has been finished (Table 46.1)
developed a postcardiotomy syndrome. Four recurrent neuro- (45) to prove superiority of the interventional versus medical
logic events occurred during a mean follow-up time of approach in patients with cryptogenic stroke. The PC-Trial with
19 months. Devuyst et al. did not see any neurologic event the Amplatzer PFO Occluder was the first one commenced and
during a two-year follow-up after surgical PFO closure despite a concluded enrollment in early 2009 with roughly 400 patients.
residual shunt in 4 of 30 patients (43). The 91 patients of Dearani It will putatively be published in 2012. In the “CLOSURE I
et al. suffered from eight neurologic events during a mean trial,” device PFO occlusion using the STARFlex Occluder is
follow-up of two years (18), 7% had a pericardial effusion, randomized to medical therapy with warfarin, acetylsalicylic
which had to be drained in 4 of 6 patients, 3% had postoperative acid, or both during 24 months of follow-up. Preliminary
bleeding, and one patient sustained superficial sternal wound results might become available in 2011. Other randomized
infection. These patients had a mean hospital stay of 6 3 days. studies are still recruiting: in the CLOSE trial, patients are
Baker compared cost implications of the interventional and the randomized either to oral anticoagulation, antiplatelet agents,
surgical approach (44). The groups were age matched, differ- or PFO closure. In the RESPECT study, PFO patients are
ences of gender, ASD size, and comorbidity were not statisti- randomized either to catheter occlusion using the Amplatzer
cally significant. He found a significant (p < 0.001) difference of PFO Occluder or best medical therapy (antiaggregation or
the length of hospital stay (two vs. five days) and of the total anticoagulation).
hospital related costs (US$7397 vs. US$15,234) with interven-
tional versus surgical techniques, respectively.
Transcatheter closure has been performed for over PFO Closure in Migraine Patients
30 years. Multiple studies have demonstrated that percutaneous Besides the prevention of cerebral ischemia, PFO closure has
ASD and PFO closure using current devices is safe and results in been shown to improve migraine in several observational
high occlusion rates. However, when comparing the most com- studies (46–48). The only randomized study of migraine
monly used devices for closure of interatrial communications, patients with PFO was the Migraine Intervention with STAR-
different complications tend to occur: in 2008, Taaffe and col- Flex Technology (MIST) trial (49). Evaluating a possible benefit
leagues published a single center randomized study comparing of PFO closure for the prevention of migraine, it has failed to
procedural complications and 30-day clinical outcomes of the show significant benefits over medical therapy. More recently,
three most common PFO closure devices with 220 patients, however, a small study by Vigna and colleagues examined a
receiving the Amplatzer PFO Occluder, the STARFlex Occluder, collective of migraine patients with large PFOs and subclinical
or the Gore Helex device (32). All PFO closures were technically brain MRI lesions. The investigators showed that in these
successful, although those performed with the Helex occlusion patients, a significantly higher reduction in the frequency and
device were more likely to require subsequent treatment with a severity of migraine recurrence can be obtained by PFO closure
different device (7 out of 220 vs. 2 out of 220 for the other two compared with control subjects (50).
groups). At 30-day follow-up, echocardiography revealed that We occluded the PFO in 17 migraine patients with the
143 (65%) of the patients who received the Amplatzer device had Amplatzer PFO Occluder (51). Complete PFO closure could be
no residual shunt, compared with 116 (53%) with the Helex achieved in 16 patients. There were no procedural complica-
device, and 137 (62%) with the STARFlex device. There was one tions and no embolic events in a mean follow-up time of
cardiac tamponade with need for surgery in the Amplatzer 2.7 years. The migraine disappeared in four patients (24%)
group, possibly due to device erosion of the aortic root. In the and improved in eight patients (47%), whereas the headaches
reported series, thrombus formation on the occluder in 8 of 220 remained unchanged in five patients (29%). The prevalence of
patients (4%) occurred in the STARFlex group only. migraine with aura decreased from 82% to 24%.
Table 46.1 Current Ongoing Trials on PFO Closure

Trial name Device Sponsor Estimated enrollment Results expected
RESPECT Amplatzer PFO Occluder AGA Medical Corporation 500 2013
CLOSURE I STARFlex Septal Closure System NMT Medical, Inc. 900 2011
PC-Trial Amplatzer PFO Occluder AGA Medical Corporation 400 2012
CLOSE Any device Assistance Publique–Hopitaux de Paris 900 2013
REDUCE Helex Septal Occluder W. L. Gore & Associates, Inc. 664 2013
Abbreviation: PFO, patent foramen ovale.
[2/7/010/10:12:57] [584–592]
CONCLUSION 18. Wahl A, Tai T, Praz F, et al. Late results after percutaneous closure
In patients with appropriate anatomical dimensions of ASD, of patent foramen ovale for secondary prevention of paradoxical
interventional closure should be favored. This technique pro- embolism using the amplatzer PFO occluder without intraproce-
vides low complication rates, a short hospitalization, and dural echocardiography: effect of device size. JACC Cardiovasc
Interv 2009; 2:116–123.
acceptable costs. Surgical ASD closure should be reserved for
19. Krumsdorf U, Ostermayer S, Billinger K, et al. Incidence and
patients with septum primum defects, sinus venosus defects, or clinical course of thrombus formation on atrial septal defect and
very large secundum defects. The significant coincidence of patient foramen ovale closure devices in 1,000 consecutive
residual leak and recurrent embolic events suggests that PFO patients. J Am Coll Cardiol 2004; 43:302–309.
closure is the right way to prevent a patient with (cryptogenic) 20. King TD, Mills NL. Nonoperative closure of atrial septal defects.
stroke and PFO from recurrences. However, randomized data Surgery 1974; 75:383–388.
are still lacking. In the last years, many new PFO devices with 21. King TD, Thompson SL, Steiner C, et al. Secundum atrial septal
different approaches for stroke prevention have entered the defect. Nonoperative closure during cardiac catheterization.
market. There is a trend toward PFO specific devices that leave JAMA 1976; 235:2506–2509.
less or no foreign material in the heart to minimize device 22. Mills NL, King TD, Joyce D. Transvenous closure of ASDs with a
double umbrella device: 7-years minimum follow-up. Circulation
related complications. Percutaneous PFO closure can be per-
1984; 70:317.
formed safely with a high success rate and low morbidity. In 23. Mills NL, King TD. Late follow-up of nonoperative closure of
observational studies, it has been shown to be effective in secundum atrial septal defects using the King-Mills double-
preventing recurrent cerebral ischemia in patients with a his- umbrella device. Am J Cardiol 2003; 92:353–355.
tory of (cryptogenic) stroke. 24. Bridges ND, Hellenbrand W, Latson L, et al. Transcatheter closure
of patent foramen ovale after presumed paradoxical embolism.
Circulation 1992; 86:1902–1908.
REFERENCES 25. Kong H, Wilkinson JL, Coe JY, et al. Corrosive behaviour of
1. Erdmann E. Klinische Kardiologie. Berlin: Springer, 2000. Amplatzer devices in experimental and biological environments.
2. Ho SY, Path FRC, McCarthy KP, et al. Anatomy of atrial and Cardiol Young 2002; 12:260–265.
ventricular septal defects. J Interv Cardiol 2000; 13:475–486. 26. Berger F, Ewert P, Bjornstad PG, et al. Transcatheter closure as
3. Hagen PT, Scholz DG, Edwards WD. Incidence and size of patent standard treatment for most interatrial defects: experience in 200
foramen ovale during the first 10 decades of life: an autopsy study patients treated with the Amplatzer Septal Occluder. Cardiol
of 965 normal hearts. Mayo Clin Proc 1984; 59:17–20. Young 1999; 9:468–473.
4. Pearson AC, Nagelhout D, Castello R, et al. Atrial septal aneurysm 27. Majunke N, Bialkowski J, Wilson N, et al. Closure of atrial septal
and stroke: a transesophageal echocardiographic study. J Am Coll defect with the Amplatzer septal occluder in adults. Am J Cardiol
Cardiol 1991; 18:1223–1229. 2009; 103:550–554.
5. Cohnheim J. Thrombose und Embolie: Vorlesung über allgemeine 28. Carminati M, Chessa M, Butera G, et al. Transcatheter closure of
Pathologie. Berlin: Hirschwald, 1877. atrial septal defects with the STARFlex device: early results and
6. Allemann Y, Hutter D, Lipp E, et al. Patent foramen ovale and follow-up. J Interv Cardiol 2001; 14:319–324.
high-altitude pulmonary edema. JAMA 2006; 296:2954–2958. 29. Mullen MJ, Hildick-Smith D, De Giovanni JV, et al. BioSTAR
7. Serena J, Davalos A. Patent foramen ovale and cryptogenic stroke: Evaluation STudy (BEST): a prospective, multicenter, phase I
where to go from here. Rev Esp Cardiol 2003; 56:649–651. clinical trial to evaluate the feasibility, efficacy, and safety of the
8. Dearani JA, Ugurlu BS, Danielson GK, et al. Surgical patent foramen BioSTAR bioabsorbable septal repair implant for the closure of
ovale closure for prevention of paradoxical embolism-related cere- atrial-level shunts. Circulation 2006; 114:1962–1967.
brovascular ischemic events. Circulation 1999; 100:II171–II175. 30. Jones TK, Latson LA, Zahn E, et al. Results of the U.S. multicenter
9. Landefeld CS, Beyth RJ. Anticoagulant-related bleeding: clinical pivotal study of the HELEX septal occluder for percutaneous
epidemiology, prediction, and prevention. Am J Med 1993; 95: closure of secundum atrial septal defects. J Am Coll Cardiol
315–328. 2007; 49:2215–2221.
10. Levine MN, Raskob G, Landefeld S, et al. Hemorrhagic compli- 31. Billinger K, Ostermayer S, Carminati M, et al. HELEX Septal
cations of anticoagulant treatment. Chest 1995; 108:276S–290S. Occluder for transcatheter closure of patent foramen ovale: multi-
11. Windecker S, Wahl A, Nedeltchev K, et al. Comparison of medical centre experience. EuroIntervention 2006; 1:465–471.
treatment with percutaneous closure of patent foramen ovale in 32. Taaffe M, Fischer E, Baranowski A, et al. Comparison of three
patients with cryptogenic stroke. J Am Coll Cardiol 2004; 44:750–758. patent foramen ovale closure devices in a randomized trial
12. Nellessen U, Daniel WG, Matheis G, et al. Impending paradoxical (Amplatzer vs. CardioSEAL-STARflex vs. Helex occluder). Am J
embolism from atrial thrombus: correct diagnosis by transesopha- Cardiol 2008; 101:1353–1358.
geal echocardiography and prevention by surgery. J Am Coll 33. Luermans JG, Post MC, Schrader R, et al. Outcome after percuta-
Cardiol 1985; 5:1002–1004. neous closure of a patent foramen ovale using the Intrasept device: a
13. Nelson CW, Snow FR, Barnett M, et al. Impending paradoxical multi-centre study. Catheter Cardiovasc Interv 2008; 71:822–828.
embolism: echocardiographic diagnosis of an intracardiac throm- 34. Ewert P, Soderberg B, Dahnert I, et al. ASD and PFO closure with
bus crossing a patent foramen ovale. Am Heart J 1991; 122:859–862. the Solysafe septal occluderresults of a prospective multicenter
14. Homma S, Sacco RL, Di Tullio MR, et al. Effect of medical treatment pilot study. Catheter Cardiovasc Interv 2008; 71:398–402.
in stroke patients with patent foramen ovale: patent foramen ovale 35. Buscheck F, Sievert H, Kleber F, et al. Patent foramen ovale using
in Cryptogenic Stroke Study. Circulation 2002; 105:2625–2631. the Premere device: the results of the CLOSEUP trial. J Interv
15. Homma S, Di Tullio MR, Sacco RL, et al. Characteristics of patent Cardiol 2006; 19:328–333.
foramen ovale associated with cryptogenic stroke. A biplane trans- 36. Sievert H, Ruygrok P, Salkeld M, et al. Transcatheter closure of
esophageal echocardiographic study. Stroke 1994; 25:582–586. patent foramen ovale with radiofrequency: acute and intermediate
16. De Castro S, Cartoni D, Fiorelli M, et al. Morphological and term results in 144 patients. Catheter Cardiovasc Interv 2009;
functional characteristics of patent foramen ovale and their 73:368–373.
embolic implications. Stroke 2000; 31:2407–2413. 37. Bogousslavsky J, Garazi S, Jeanrenaud X, et al. Stroke recurrence
17. Mas JL, Arquizan C, Lamy C, et al. Recurrent cerebrovascular in patients with patent foramen ovale: the Lausanne Study.
events associated with patent foramen ovale, atrial septal aneur- Lausanne Stroke with Paradoxal Embolism Study Group. Neurol-
ysm, or both. N Engl J Med 2001; 345:1740–1746. ogy 1996; 46:1301–1305.
[2/7/010/10:12:57] [584–592]
38. Comess KA, DeRook FA, Beach KW, et al. Transesophageal 45. O’Gara PT, Messe SR, Tuzcu EM, et al. Percutaneous device
echocardiography and carotid ultrasound in patients with cerebral closure of patent foramen ovale for secondary stroke prevention:
ischemia: prevalence of findings and recurrent stroke risk. J Am a call for completion of randomized clinical trials. A science
Coll Cardiol 1994; 23:1598–1603. advisory from the American Heart Association/American Stroke
39. Mas JL, Zuber M. Recurrent cerebrovascular events in patients Association and the American College of Cardiology Foundation.
with patent foramen ovale, atrial septal aneurysm, or both and J Am Coll Cardiol 2009; 53:2014–2018.
cryptogenic stroke or transient ischemic attack. French Study 46. Reisman M, Christofferson RD, Jesurum J, et al. Migraine headache
Group on Patent Foramen Ovale and Atrial Septal Aneurysm. relief after transcatheter closure of patent foramen ovale. J Am Coll
Am Heart J 1995; 130:1083–1088. Cardiol 2005; 45:493–495.
40. Stone DA, Godard J, Corretti MC, et al. Patent foramen ovale: 47. Schwerzmann M, Wiher S, Nedeltchev K, et al. Percutaneous
association between the degree of shunt by contrast transesopha- closure of patent foramen ovale reduces the frequency of migraine
geal echocardiography and the risk of future ischemic neurologic attacks. Neurology 2004; 62:1399–1401.
events. Am Heart J 1996; 131:158–161. 48. Tobis MJ, Azarbal B. Does patent foramen ovale promote cryp-
41. Khairy P, O’Donnell CP, Landzberg MJ. Transcatheter closure togenic stroke and migraine headache? Tex Heart Inst J 2005;
versus medical therapy of patent foramen ovale and presumed 32:362–365.
paradoxical thromboemboli: a systematic review. Ann Intern Med 49. Dowson A, Mullen MJ, Peatfield R, et al. Migraine Intervention
2003; 139:753–760. With STARFlex Technology (MIST) trial: a prospective, multi-
42. Devuyst G, Bogousslavsky J, Ruchat P, et al. Prognosis after stroke center, double-blind, sham-controlled trial to evaluate the effec-
followed by surgical closure of patent foramen ovale: a prospec- tiveness of patent foramen ovale closure with STARFlex septal
tive follow-up study with brain MRI and simultaneous trans- repair implant to resolve refractory migraine headache. Circula-
esophageal and transcranial Doppler ultrasound. Neurology tion 2008; 117:1397–1404.
1996; 47:1162–1166. 50. Vigna C, Marchese N, Inchingolo V, et al. Improvement of
43. Homma S, Di Tullio MR, Sacco RL, et al. Surgical closure of patent migraine after patent foramen ovale percutaneous closure in
foramen ovale in cryptogenic stroke patients. Stroke 1997; 28: patients with subclinical brain lesions: a case-control study.
2376–2381. JACC Cardiovasc Interv 2009; 2:107–113.
44. Baker SS, O’Laughlin MP, Jollis JG, et al. Cost implications of 51. Wahl A, Praz F, Findling O, et al. Percutaneous closure of patent
closure of atrial septal defect. Catheter Cardiovasc Interv 2002; foramen ovale for migraine headaches refractory to medical treat-
55:83–87. ment. Catheter Cardiovasc Interv 2008; 74:124–129.
[22/6/010/11:53:2] [593–601]
47
Pediatric and adult congenital cardiac interventions

Sawsan M. Awad, Qi-Ling Cao, and Ziyad M. Hijazi
INTRODUCTION Percutaneous VSD Closure

Interventional cardiology is a rapidly progressing field that has Anatomy/embryology considerations. VSDs are considered the
seen major growth in the last three decades. Interventional most common cardiac abnormalities found in children,
therapy has become an acceptable alternative treatment and a accounting for approximately 30% of all defects (1,2).
standard of care for many pediatric and adult patients with VSDs may be located in four different positions according
congenital heart disease (CHD). In cases where conventional to the Kirklin classification (3). Membranous VSDs are the most
interventional therapy is not feasible due to limited access or common (representing 75%), followed by the muscular type,
associated conditions that may require surgical repair, a hybrid which represents 10% to 15%. Rare acquired causes of VSD
approach has been increasingly implemented with obvious include a defect following traumatic injury to the chest or
advantages to the patient. This concept involves direct expo- myocardial infarction. Acquired VSDs are encountered in
sure of the heart via a median sternotomy performed by the 0.2% of all myocardial infarction patients.
surgeon, followed by an interventional procedure performed Approximately 70% of muscular VSDs and 30% of mem-
by the team without subjecting the patient to cardiopulmonary branous VSDs close spontaneously within five years after initial
bypass. diagnosis. Patients with large defects (as big as the aortic valve)
In this chapter, we will discuss the congenital cardiovas- may present with signs and symptoms of congestive heart
cular conditions for which the interventional therapy has been failure or failure to thrive. All VSDs can be repaired surgically
established as the preferred management strategy, including with the exception of apical defects and the Swiss cheese type
atrial septal defect, patent foramen ovale, and ventricular septal of VSD. The overall risk for VSD surgical repair is <5%.
defect (VSD) device closure, as well as semilunar valvuolo- Mortality and morbidity rates increase with multiple VSDs,
plasty for congenital aortic and pulmonary valve stenoses. In pulmonary hypertension, residual VSD, and complex associ-
addition, we will summarize the value of interventional thera- ated anomalies. Complete heart block immediately after
pies for selected adult congenital cardiac lesions, such as surgery occurs in <1% of patients. Late-onset complete heart
coarctation of the aorta (COA), pulmonary artery stenosis block is a rare problem, especially in patients with postopera-
(PAS), coronary artery fistula (CAF), aortopulmonary collater- tive complete right bundle branch block and left anterior
als (APCs), pulmonary arteriovenous fistula, and finally the hemiblock.
emerging percutaneous pulmonary valve implantation for Lock and colleagues (4) performed the first percutaneous
patients with severe pulmonary insufficiency and/or stenosis. VSD device closure in 1987, and since then many devices have
been used for this purpose. The most widely used device in the
United States and Europe is the Amplatzer muscular VSD
EQUIPMENT occluder device (AGA Medical, Plymouth, Minnesota, U.S.). It
The performance of interventional cardiac catheterization pro- is specifically designed for the ventricular septum and is made
cedures for CHD, both in adult and pediatric patients, requires of 0.004- to 0.005-in. nitinol wire. Holzer and colleagues
well-trained cardiologists and a well-equipped cardiac cathe- reported on the use of this device in a U.S. registry with good
terization laboratory. A description of training requirements for results (5). In that registry, a total of 75 patients were enrolled in
interventionalists performing the procedures is beyond the 14 cardiac centers in the United States. The median age of
scope of this chapter. Biplane fluoroscopy imaging is preferred patients was 1.4 years (range 0.1–54.1 years). The median size
as it allows a better understanding of the anatomy and location of the primary VSD was 7 mm (range 3–16 mm). Implantation
of the defect and minimizes the amount of contrast agent of multiple devices was required in 20.5% of the procedures.
administered. Surgical backup and extracorporeal membrane The complete closure rate immediately after the procedure
oxygenator support capabilities are mandatory in any institu- was 47.2%, increasing to 69.6% at 6 months, and 92.3% at
tion planning to have an interventional program for CHD. 12-month follow-up. Major complications occurred in 10.7%
Since an interventional procedure for CHD should never fail of patients; however, only 2.7% of patients died due to the
due to lack of proper equipment, multiple types of wires, procedure. The percutaneous closure approach is preferred for
catheters, balloons, retrieval catheters or snares, and devices children larger than 5 kg. However, if the patient’s weight is
should be readily available. <5 kg or if the septal anatomy precludes a percutaneous
approach [e.g., transposition of the great arteries, double outlet
right ventricle (RV)], or if the venous access is not present, the
INTERVENTIONS hybrid approach has been advocated as an alternative, without
Atrial septal defect and patent foramen ovale closure are cardiopulmonary bypass (6). This involves collaboration
discussed in chapter 46 and are not further discussed here. between the surgeons and the interventional cardiologists.
[22/6/010/11:53:2] [593–601]
For single muscular VSD, the procedure can be safely pulmonary artery it is snared and exteriorized out into the
performed without transesophageal echocardiographic (TEE) femoral vein. The delivery sheath is then placed over this wire
guidance; however, for multiple defects/Swiss cheese septum, into the LV apex. The remaining steps are similar to the above.
it is advisable to perform the procedure under TEE guidance. After device release, a brief complete TEE study is performed
Femoral artery and vein are accessed routinely. If the VSD is with additional imaging in multiple planes to confirm device
located in the mid, posterior, or apical septum, the right inter- placement and to assess for residual shunting or any obstruc-
nal jugular vein is also accessed. The patient is heparinized to tion or regurgitation induced by the device. The device orien-
achieve an activated clotting time of >200 seconds at the time of tation commonly changes slightly to align with the septum as it
device placement. Routine right and left heart catheterization is is released from the delivery cable and all tension on the device
performed to assess the degree of shunting and to evaluate the is eliminated. Additional VSDs are then occluded in the same
pulmonary vascular resistance. Angiography in single plane fashion. Repeat LV angiogram in 358 LAO/358 cranial view is
(358 LAO/358 cranial) is performed to define the location, size, performed 10 minutes after final device release to assess the
and number of VSDs. This projection profiles the muscular result. Patients receive a dose of an appropriate antibiotic
septum. A complete TEE study is performed. Specific attention (commonly Cefazolin at 20 mg/kg) during the catheterization
is paid to the characterization of the VSDs and nearby struc- procedure and two further doses at eight-hour intervals. The
tures, including the papillary muscles, moderator band, and the patients are recovered in an appropriate setting (usually inten-
chordae tendonae. The atrioventricular valves are interrogated sive care unit) and are routinely discharged home the following
at baseline for any regurgitation. The appropriate Amplatzer day. Observation of subacute bacterial endocarditis prophy-
VSD device size is chosen to be 1 to 2 mm larger than the VSD laxis is recommended for six months or until complete closure
size as assessed by TEE or angiography at end diastole (the is obtained. Patients are instructed to avoid contact sports for
bigger of the two diameters). The authors depend mainly on one month. Follow-up includes TTE, chest radiograph, and
echocardiographic and/or angiographic sizing of the defect electrocardiogram at six months post closure and yearly there-
while others rely on balloon-sizing. A long venous sheath after. Figure 47.1 demonstrates closure steps in a patient with
(6–8 Fr) is then placed across the VSD. This is accomplished in a VSD.
variety of ways. The most common approach used for the mid-
muscular VSDs is to advance a curved 4-Fr end-hole catheter
(Judkins right or Cobra) from the left ventricle (LV) across the Congenital Aortic Stenosis/Pulmonary Stenosis
VSD into the RV. An exchange-length 0.035-in. J-tipped guide- Valvular aortic stenosis (AS) occurs in approximately 3% to 6%
wire is advanced through the VSD and the RV into either of patients with CHD (2). The stenotic valve is usually secondary
pulmonary artery branch. This wire is then snared and exteri- to aortic valve maldevelopment with increased thickening and
orized through the right internal jugular vein. This provides a rigidity of the valve tissue and variable degrees of commissural
stable arteriovenous loop and allows a 6- to 8-Fr long Mullins fusion. Compensatory left ventricular hypertrophy is propor-
type sheath (AGA Medical) to be advanced from the jugular tional to the degree of obstruction. With severe hypertrophy and
vein to the RV and positioned into the LV. The approach from valvar obstruction, myocardial ischemia may result from the
the jugular vein provides a straight course for mid, apical, or combination of limited cardiac output, reduced coronary perfu-
posterior defects. Some larger mid-muscular or apical VSDs can sion, and increased myocardial oxygen consumption.
be easily crossed from the RV side. However, care should be In neonatal critical AS, congestive heart failure and shock
exercised not to go through the trabeculae in the RV. Once a occurs around the time of natural patent ductus arteriosus clo-
catheter crosses into the LV, an exchange-length guidewire is sure. In older children and adolescents, the presentation could be
positioned into the LV apex and a 6- to 8-Fr long Mullins type the systolic ejection murmur characteristic for valvar AS.
sheath is advanced over this wire and positioned into the body Balloon aortic valvuloplasty is a safe initial treatment in
of the LV. On occasion, after removal of the dilator and wire most patients with congenital aortic valve stenosis. Patients
from the long sheath, kinking of the distal part of the sheath (at with severely dysplastic valves may have less favorable results
the ventricular septum) is encountered. In such circumstances, with balloon aortic valvuloplasty, but in most patients the
a 0.018-in. J-tipped glidewire is advanced through the dilator results are similar to those obtained with surgical valvotomy
and left inside the sheath while advancing the device beside it. (7,8). The overall goal, especially in neonates and infants, is to
This helps minimize kinking of the sheath. Once the device relieve the aortic valve obstruction sufficiently without devel-
approaches the tip of the sheath, this wire is removed and the opment of significant valve insufficiency, thereby resulting in
device is deployed in the usual fashion. TEE and angiography normalization of left ventricular systolic function. Achievement
using a pigtail catheter positioned in the LV are very helpful of this goal typically entails performing a conservative balloon
imaging techniques in guiding device position. The LV disk is valvuloplasty by reducing the peak-to-peak systolic gradient by
deployed in the middle of the LV, and the entire assembly 50%. Balloon diameters are usually 85% to 90% of the aortic
(cable/sheath) is pulled into the VSD with further retraction of valve annulus dimension measured via aortic angiography. If
the sheath to expand the waist inside the septum. Repeat TEE unsatisfactory result is encountered and no significant increase
and angiography to confirm optimal device position prior to in aortic regurgitation is noticed, a higher balloon size can be
deployment of the RV disk is of great importance. Once posi- used to repeat the procedure. In critically ill patients, surgical
tion is confirmed, further retraction of the sheath to expand the backup and circulatory support in the form of an extracorpor-
RV disk is performed. Again, prior to device release, repeat eal membrane oxygenator should be available.
TEE and angiography are performed. If device position is Right and left heart catheterization assessment is indi-
satisfactory, the device is released by counterclockwise rotation cated prior to valvuloplasty to evaluate the right-sided pressure
of the cable using the pin vise. For anterior muscular VSDs, we and the left ventricular end-diastolic pressure. It is best to
prefer the femoral vein approach as it is an easier course for measure the gradient across the valve by placing a catheter
the delivery sheath. Once the wire crosses the VSD into the above the valve and another one in the LV (simultaneous). In
[22/6/010/11:53:2] [593–601]
PEDIATRIC AND ADULT CONGENITAL CARDIAC INTERVENTIONS 595
Figure 47.1 Cine fluoroscopic images during ventricular septal defect (VSD) closure. (A) Angiogram in the left ventricle in the long axial
oblique view demonstrating the presence of high muscular VSD (arrow). (B) Cine fluoroscopy after the defect has been crossed and the wire
was exteriorized from the right femoral vein and the delivery sheath (arrow) was positioned in the left ventricle. Note the arteriovenous loop is
still established. (C) The delivery sheath with the Amplatzer device (arrow) is in mid left ventricle. (D) Deployment of the left desk (arrow) in the
left side of the defect. (E) The device was released from the cable (arrow). (F) Final left ventricle angiogram showing complete closure of the
defect.
symptomatic patients, this condition may be treated by valvu- Patients with mild PS are usually asymptomatic. The
loplasty even in the presence of a lower gradient. In asympto- diagnosis is usually made during routine physical examination
matic patients, a peak-to-peak gradient of over 55 mmHg with audible ejection systolic murmur. Patients with moderate
is an indication to proceed. However, often under general or severe degree of PS may have mild exertional dyspnea.
anesthesia, the gradient is lower than under normal conditions. Adults may be asymptomatic irrespective of the severity of
Therefore, prior assessment by echocardiography or the admin- their obstruction. Patients with severe PS present with signs of
istration of dobutamine during catheterization will unmask congestive heart failure and cyanosis due to shunting of blood
significant obstruction. Patients with an aortic gradient across patent foramen ovale or atrial septal defect. Treatment is
<55 mmHg should be followed medically unless they become indicated in asymptomatic patients with severe PS (peak-to-
symptomatic. peak gradient >55 mmHg) or in symptomatic patients with
Retrograde crossing of the aortic valve via the femoral evidence of right ventricular dysfunction irrespective of the
artery is preferred; however, in neonates, a carotid cut-down is gradient.
employed by many interventionalists to allow crossing of the Balloon pulmonary valvuloplasty was initially described
valve. Trans-septal approach is another technique to cross the by Kan et al. (10). The success rates are excellent (85%) for
valve in an antegrade fashion that was reported to decrease the children with classical PS; however, for patients with dysplastic
risk of silent cerebral embolism frequently seen with the retro- valves or supravalvular and/or subvalvular PS, success rate is
grade approach at a rate of 22% (9). low (35–65%).
Left ventriculography demonstrates the stenotic valve In selected infants with membranous pulmonary atresia
orifice. It is also helpful in evaluating the subaortic and supra and adequate size RV, perforation of the membrane with either
aortic areas for the presence of an additional level of stenosis. a stiff end of a wire or radiofrequency perforation catheter
Selection of the balloon size depends on the orifice size followed by balloon dilation has been successful in creating an
measured during angiographic assessment. The chosen balloon open right ventricular outflow tract (11–14).
should be 85% to 90% of the size of the stenotic valve to Femoral venous access and right heart catheterization to
minimize the risk of regurgitation. Figure 47.2 demonstrates assess the hemodynamics are performed. Femoral artery access
TEE images in an adolescent with valvar AS who underwent is rarely needed. In neonatal critical PS, umbilical vein access
successful balloon valvuloplasty. may be used. Right ventricular angiography is performed in
Isolated valvar pulmonary stenosis (PS) represents 8% to both anteroposterior and lateral projections to identify the
10% of all patients with CHD. The stenotic valve is usually pulmonary valve and measure the annular size (between the
dome shaped, with diffuse thickening and commissural fusion. hinge points). The balloon size is selected to be no more than
[22/6/010/11:53:2] [593–601]
Figure 47.2 (See color insert) Transesophageal echocardiographic images during balloon aortic valvuloplasty in a teenage boy.
(A) Longitudinal view demonstrating the domed stenotic aortic valve; (B) same view as A with color Doppler showing the stenotic mosaic
color jet; (C) balloon inflation; (D) after the valvuloplasty showing complete opening of the valve; (E) with color during systole showing
improved opening of valve; and (F) diastolic frame showing no aortic insufficiency. Abbreviations: LA, left atrium; LV, left ventricle; RV, right
ventricle; AO, aorta.
1.2 times the diameter of the valve annulus. A balloon end-hole Indications for COA treatment are basically the same as
catheter is advanced to the distal right or left pulmonary artery. those for surgery and include hypertension proximal to the
A stiff exchange-length guidewire is then placed in the branch coarctation with a resting systolic pressure gradient across the
pulmonary artery. We prefer to insert the balloon via a sheath. narrowed segment >20 mmHg or angiographically severe
This we believe minimizes access vessel injury. The balloon is coarctation with extensive collaterals (15).
introduced over the guidewire and is centered at the pulmo- Coarctation angioplasty requires femoral venous and
nary valve. Adjustment of the balloon position may be per- arterial access. The patient is heparinized to achieve an acti-
formed by repeated small pressure inflations and waist vated clotting time of >200 seconds. Retrograde approach is the
verification. The balloon is inflated rapidly until the waist most commonly used technique; however, on occasions in
disappears then it is deflated immediately. If suboptimal results severe coarctation or if there is acquired atresia, crossing
are obtained, repositioning of the balloon and repeating the from above (trans-septal or left subclvian artery) is indicated.
previous steps may be done. Larger balloon size may be used Right and left heart catheterization with hemodynamic assess-
for the second inflation if optimal results are not achieved and ment is performed. Biplane aortography is performed. The
no pulmonary regurgitation is observed. narrowest area of the coarctation is measured as well as the
proximal, distal, and the aorta at the level of the diaphragm.
Balloon size selection depends on the nature of the coarctation.
INTERVENTION IN ADULTS WITH CHD In native coarctation, the balloon size is selected to be equal to
Coarctation of the Aorta the aorta at the isthmus. In postoperative coarctation, the bal-
COA is a relatively common defect that accounts for 5% to 8% loon is chosen to be around 2.5 to 3 times the diameter of the
of all CHDs. COA occurs as an isolated lesion or in association narrowest coarctation area but not more than the area imme-
with other congenital heart defects, the most common of which diately distal to the coarctation (post-stenotic dilatation).
are bicuspid aortic valve and VSD. The selected balloon is centered at the coarctation area.
Coarctation almost always occurs in the thoracic aorta The balloon is inflated until the waist disappears and then
distal to the origin of the left subclavian artery. deflated immediately. Simultaneous pressure measurement
Neonates with critical COA may present with shock-like between the ascending and descending aorta pressures is
symptoms around the time of ductal closure. Less severe COA obtained. Angiography is performed after balloon inflation to
may not be detected until later in childhood or even in adult- determine the effect of angioplasty on the coarctation area.
hood. One of the most common late presentations is systemic Tears or dissection of the aorta may be also detected. If optimal
hypertension. Heart murmur may be another late presentation results are not achieved, a larger balloon may be used for a
sign secondary to the presence of a well-developed collateral second inflation. Temporary chest pain during balloon inflation
arterial system. is acceptable. If chest pain persists after balloon deflation,
[22/6/010/11:53:2] [593–601]
Figure 47.3 Cine fluoroscopic images in straight lateral

projection in a patient with coarctation of the aorta.
(A) Angiogram in the ascending aorta showing severe
coarctation of the aorta (arrow) and presence of collaterals.
(B) Cine frame showing the stent that was implanted at the
site of coarctation. (C) Final angiogram showing good stent
position, no residual obstruction. The distal end of the stent
is not fully apposed to the wall of the aorta.
dissection may be present. It is important to note that at no time Despite advances in balloon types and pressure achieved,
is the freshly dilated area crossed with a catheter. Therefore, around one-third of vessels, more often distal, are resistant to
once a guidewire is left in place and is used for the angioplasty, angioplasty. The use of cutting balloons is an effective treat-
all catheter manipulations are performed over this wire to ment for small lobar PAS refractory to balloon angioplasty (17).
prevent disruption of the freshly dilated area. Stenting of branch pulmonary arteries is frequently used
In the adult patient, stent angioplasty is currently pre- in children with PAS and/or hypoplasia (18–20). Because of the
ferred over balloon angioplasty for both native and recurrent higher immediate success and less incidence of restenosis,
coarctation. Stent placement eliminates the elastic recoil of aortic stenting of pulmonary arteries may be a reasonable first-line
tissue observed after balloon angioplasty. It also allows the use therapy.
of smaller balloon size, minimizing the risks of using large The most commonly used access is the femoral vein.
balloons. Stent placement was noticed to have less residual Right heart catheterization with hemodynamic assessment is
gradient and better opening of the coarctation segment. Further, performed first. Pulmonary artery angiography is then per-
we believe that covered stents should be available in any catheter- formed to localize the affected segment(s). Selective injection to
ization laboratory engaged in balloon/stent angioplasty of coarc- the lung and the lobe affected is highly recommended. Before
tation. We routinely use covered stents for appropriate coarctation starting the balloon dilation procedure, a stiff exchange-length
in any patient with an adult-size aorta (around 18–20 mm). wire should be placed in a large vessel distal to the stenotic
Covered stents can be used as a primary tool or as bailout therapy branch.
after angioplasty to treat dissections or aneurysms. Figure 47.3 The selected balloon diameter should be two to four
demonstrates a patient with tight coarctation who underwent times the diameter of the narrow segment but not more than
successful stent placement. two times the diameter of the normal vessel on either side of
the lesion (21).
Pulmonary Artery Stenosis The balloon is inflated until the waist disappears. After
PAS accounts for 2% to 3% of all CHD. PAS may be isolated dilation, the area is reassessed by pressure recording across the
(Williams syndrome, Alagille syndrome) or associated with dilated area and angiography. Successful dilation results in
more complex CHD, such as tetralogy of Fallot or transposition improved diameter of the segment, increase in distal pressure,
of the great arteries. PAS may be also central, peripheral, and/or a decrease of >20% in systolic right ventricular to aortic
intermediate, unilateral, or bilateral. pressure ration.
Gay et al. established four PAS classes according to the Cutting balloons may be used for vessels resistant to
anatomic location of the narrowing. Type I includes single conventional balloon angioplasty. Initially, the vessel is dilated
constriction of varying length confined to the main pulmonary with the cutting balloon then a high-pressure balloon is
artery (MPA), the right pulmonary artery (RPA), or the left employed to open up the vessel to a larger diameter. One
pulmonary artery (LPA) (types IA, IB, and IC, respectively). limiting factor about cutting balloons is the lack of availability
Type II includes bifurcation stenosis where the constriction of large sizes, since the largest cutting balloon available is
involves the distal end of the MPA, RPA, and LPA. This type 8 mm. Therefore, cutting balloon angioplasty is good for vessels
is further subdivided into IIA where the narrowing segment is up to 8 mm in diameter.
short and localized and IIB where the narrowing segment is Stent placement of the affected area will avoid the recoil
long. Type III includes multiple peripheral stenoses, and type nature encountered after balloon angioplasty. However, stent
IV includes combined central and peripheral stenoses (16). placement is technically more challenging, and potentially
Patients with mild to moderate arterial obstruction are could have more complications and could commit the patient
usually asymptomatic. Cases with severe stenosis may have to repeat interventions, especially if continued growth of the
dyspnea on exertion, easy fatigability, and occasional right- patient is expected. Figure 47.4 demonstrates a patient who
sided heart failure. Treatment of PAS depends on the site of the underwent placement of two stents to treat bilateral stenoses at
stenotic segment. Percutaneous pulmonary angioplasty is suit- the origin of the branch pulmonary arteries from the MPA.
able for distal lesions unreachable by surgery, and surgical Intraoperative intravascular stent placement is a hybrid
pulmonary arterioplasty is feasible for more proximal lesions. technique that could be used in difficult situations. It is suitable
Variable modalities of percutaneous techniques include in the early postoperative period, difficult vascular anatomy,
balloon pulmonary angioplasty using high-pressure balloons, bilateral stenoses requiring simultaneous stent implantation,
cutting balloon angioplasty, and intravascular stent placement. short proximal segment stenosis (too short for the shortest stent
[22/6/010/11:53:2] [593–601]
Figure 47.4 Cine fluoroscopic views in a patient with bilateral branch pulmonary artery stenoses at their origin. (A) Angiogram in the main
pulmonary artery showing the narrowing at the origin of each branch (white and black arrows). (B) Stent deployment in the left pulmonary
artery (arrow). (C) Angiogram in the left pulmonary artery after the stent deployment showing expansion of the area with good result. (D) Stent
deployment in the right pulmonary artery (arrow). (E) Angiogram in the right pulmonary artery showing expansion of the area and good result.
(F) Final angiogram showing elimination of the narrowing.
available), marginal hemodynamics, severe bilateral branch Cardiac catheterization is the main diagnostic technique.
stenosis, patients on extracorporeal membrane oxygenator, Cardiac catheterization is needed initially to assess the hemo-
and patients with other cardiac lesions requiring concomitant dynamic significance of the fistula and to provide detailed
surgery (22). anatomy including size, origin, course, presence of any steno-
sis, and the drainage site.
The main goal of treatment of CAF is complete occlusion
Coronary Artery Fistula with no residual fistulae. Catheter closure of the fistulas is now
CAF is a connection between one or more of the coronary arteries considered to be a safe and effective alternative to surgery.
and a cardiac chamber or great vessel, bypassing the myocardial Catheter closure should be as distal to the end point of the
capillary bed. The abnormality is rare—the exact incidence is fistula as possible to avoid possible occlusion of branches to the
unknown—and usually occurs as an isolated finding. normal myocardium.
The fistulas originate from the right coronary artery in Selective coronary angiography is needed to confirm the
more than half of the cases. The left anterior descending coro- diagnosis and the detailed anatomy of the fistula. Detailed
nary artery is the next most frequently involved, in approxi- angiographic views in multiple projections are essential to the
mately one-third of cases, followed by the circumflex coronary successful treatment of these fistulas.
artery (23). Most of the fistulas from either coronary artery Access is usually obtained in both femoral arteries and
drain into the right side of the heart. The RV is the most one femoral vein. One artery is used for angiographic assess-
common site for drainage followed by the right atrium, coro- ment and the other for the actual closure of the fistula. We use a
nary sinus, and lastly the pulmonary artery trunk. coaxial system to manage these fistulas. A coronary guide
Most patients with CAF are asymptomatic. Diagnosis is catheter of proper size and shape is advanced to the ostium
usually suspected when a continuous murmur is detected of the involved coronary artery. The fistula is crossed with the
during a routine visit or examination for other reasons. Symp- proper coronary exchange-length wire. Then a Berman end-
toms depend on the size of the fistula, which is usually small, hole balloon catheter is passed over this wire inside the guide
and the pressure difference between the two sides of the fistula. catheter, and the balloon is positioned distal to the last viable
Rarely, congestive heart failure occurs. CAFs usually are small myocardial branch and is inflated with contrast to temporarily
in size and close spontaneously by the second decade of life. If occlude the vessel for 5 to 10 minutes to assess the risk of
not closed, complications may be the first presenting symptom. ischemia with fistula occlusion. If no detectable ischemic
Reported complications include steal from the adjacent myo- changes are noted, then the choice of technique is based on
cardium causing myocardial ischemia, thrombosis and embo- the size of the fistula and on the available equipment. Coils or
lism, cardiac failure, atrial fibrillation, rupture, endocarditis/ devices (Amplatzer PDA device, Amplatzer vascular plug,
endarteritis, and arrhythmias (23–26). Other reported rare Amplatzer VSD devices) can be used to close the fistula. If
complications include thrombosis within the fistula, leading coils are chosen, they are usually deployed retrograde (going
to acute myocardial infarction; atrial or ventricular arrhyth- from the guide catheter inside the delivery catheter positioned
mias; and spontaneous rupture of the aneurysmal fistula, distal) and if devices are chosen, usually, the wire is snared and
causing hemopericardium (27,28). exteriorized from either femoral or jugular vein and the proper
[22/6/010/11:53:2] [593–601]
univentricular physiology include left-to-right shunts causing

volume overload on the ventricle and unwanted blood return
on the operative field during cardiopulmonary bypass. The
increased pulmonary blood flow also raises the pulmonary
artery pressure. Ventricular volume overload and elevated
pulmonary artery pressure have been identified as risk factors
for the Fontan operation. Hemoptysis from rupture of thin-
walled vessels may also occur as one of the adverse effects of
APCs.
Before an aortopulmonary collateral artery is embolized,
the presence of dual blood supply from the true pulmonary
arteries to the affected segment of the lung should be deter-
mined. Embolization should only be performed if true pulmo-
nary arteries supply the same territories as the collateral vessel.
Various agents are used for embolization of APCs,
including tissue adhesives, Gelfoam, detachable balloons, and
coils. Sharma et al. reported on APC occlusions in one of the
largest series in this field (31). Indications for occlusion in this
study were hemoptysis, intractable cardiac failure, and routine
preoperative procedure in most of the patients. Complete
occlusion was achieved in 76% of cases with an 8% failure
rate secondary to migration of the coils to the distal pulmonary
arteries.
The technique of closure is similar to the techniques used
Figure 47.5 Cine fluoroscopic images in a patient with left anterior in CAF embolization and other extracardiac vascular abnor-
descending coronary artery to right ventricle fistula. (A) Selective left malities. Coil migration is still one of the most common
main coronary angiogram showing the distal end of the fistula complications and use of detachable coils leads to a marked
draining into the right ventricle. (B) The fistula was crossed decrease in the risk of coil migration.
(arrow) retrograde and the wire was exteriorized from the right
internal jugular vein. (C) Deployment of the second Amplatzer
vascular plug from the right internal jugular vein to the distal end Pulmonary Arteriovenous Fistula
of the fistula, note the two plugs (arrows). (D) Final angiogram Pulmonary arteriovenous fistula is an abnormal communica-
showing complete closure of the fistula. tion between pulmonary arteries and pulmonary veins. It can
be either congenital (in most of the cases) or sometimes
acquired, especially in patients after bidirectional Glenn anas-
tomosis for single ventricle.
The clinical presentation varies from no symptoms to
size sheath is advanced into the fistula and then the device is severe illness. The most common clinical presentation includes
advanced from the vein to the fistula. The guide catheter is epistaxis, dyspnea, and hemoptysis. Cyanosis may occur from
used for selective injections to guide the deployment site and to the right-to-left shunt between the pulmonary arteries and
look for other feeding vessels (29). pulmonary veins. The resultant shunt may lead to paradoxical
Complete occlusion of the fistula may be achieved in embolus resulting in stroke or transient ischemic attack.
>95% of patients. The remaining patients may be managed Acquired fistula is seen in patients after cavopulmonary anas-
conservatively if the residual fistulas are small, or further tomosis and less in patients after Fontan surgery. Transcatheter
procedures may be required. embolization of pulmonary arteriovenous fistula is considered
Complications of the procedure are rare and include coil the main stay of treatment of this abnormality (32–35). The
embolization, transient ischemic changes, transient bundle technique of embolization is not different from closure of the
branch block, and myocardial infarction (30). Post closure, we other extracardiac vascular anomalies (CAFs and APCs). The
recommend the use of antiplatelet agents (if fistulas are small) most commonly used device is the Gianturco coils. Emboliza-
or anticoagulation (if large) to prevent thrombosis and closure tion of the entire nidus of the malformation is important to
of the larger main coronary artery. Figure 47.5 demonstrates a prevent recanalization. All feeding vessels should be identified
patient with large left anterior coronary artery to RV fistula and embolized. The risk of coil migration through the malfor-
who underwent closure from the right internal jugular vein mation into systemic circulation has been eliminated with
using two vascular plugs. detachable coils. Detachable balloons are reserved for malfor-
mations with larger feeding vessels >1 cm in diameter.
Aortopulmonary Collaterals
Collateral vessels occur in a wide variety of conditions, includ- Percutaneous Pulmonary Valve Implantation
ing pulmonary atresia with VSD, tetralogy of Fallot, scimitar The presence of severe pulmonary insufficiency may lead to RV
syndrome, and in hearts undergoing Fontan-type reconstruc- enlargement and dysfunction. Surgical placement of a compe-
tion. They may be arterial or venous and may shunt left-to-right tent valve between the RV and pulmonary arteries to treat this
or right-to-left. APCs typically originate from the descending condition requires cardiopulmonary bypass, which may aggra-
aorta and connect with true pulmonary arteries at different vate an already compromised RV. Bonhoeffer was the first to
levels. The potential disadvantage of APCs in patients with place a valve in this position percutaneously (36) and since then
[22/6/010/11:53:2] [593–601]
growing and expanding field that now involves at its far end
percutaneous valve placement and valve repair. In the near
future, further major advances in this field are likely.
REFERENCES
1. Lewis DA, Loffredo CA, Corre-Villasenor A, et al. Descriptive
epidemiology of membranous and muscular ventricular septal
defects in the Baltimore-Washington Infant Study. Cardiol Young
1996; 6:281–290.
2. Miyague NI, Cardoso SM, Meyer F, et al. Epidemiological study of
congenital heart defects in children and adolescents. Analysis of
4,538 cases. Arq Bras Cardiol 2003; 80:269–278.
3. Kirklin JW, Barrat-Boyes BG. Cardiac surgery: morphology, diag-
nostic criteria, natural history, techniques, results and indications.
In: Kirklin JW, Barrat-Boyes BG, eds. Ventricular Septal Defects.
New York: Churchill Livingstone, 1985:749–824.
4. Lock JE, Block PC, McKay RG, et al. Transcatheter closure of
ventricular septal defects. Circulation 1988; 78:361–368.
5. Holzer R, Balzer D, Cao QL, et al. Device closure of muscular
ventricular septal defects using the Amplatzer muscular ventric-
ular septal defect occluder: immediate and mid-term results of a
U.S. registry. J Am Coll Cardiol 2004; 43:1257–1263.
6. Bacha EA, Cao QL, Galantowicz ME, et al. Multicenter experience
with perventricular device closure of muscular ventricular septal
defects. Pediatr Cardiol 2005; 26:169–175.
7. Moore P, Egito E, Mowrey H, et al. Midterm results of balloon
dilation of congenital aortic stenosis: predictors of success. J Am
Coll Cardiol 1996; 27:1257–1263.
8. Pedra CA, Sidhu R, McCrindle BW, et al. Outcomes after balloon
dilation of congenital aortic stenosis in children and adolescents.
Cardiol Young 2004; 14:315–321.
9. Omran H, Schmidt H, Hackenbroch M, et al. Silent and apparent
cerebral embolism after retrograde catheterisation of the aortic
valve in valvular stenosis: a prospective, randomised study.
Lancet 2003; 361:1241–1246.
10. Kan JS, White RI Jr., Mitchell SE, et al. Percutaneous transluminal
Figure 47.6 (A) Medtronic Melody valve (Medtronic, Minneapolis, balloon valvuloplasty for pulmonary valve stenosis. Circulation
Minnesota, U.S.); (B) Edwards SAPIEN THV (Edwards Lifescien- 1984; 69:554–560.
ces, Irvine, California, U.S.). 11. Alwi M, Geetha K, Bilkis AA, et al. Pulmonary atresia with intact
ventricular septum percutaneous radiofrequency-assisted valvot-
omy and balloon dilation versus surgical valvotomy and Blalock
Taussig shunt. J Am Coll Cardiol 2000; 35:468–476.
12. Hausdorf G, Schneider M, Lange P. Catheter creation of an open
hundreds of patients have benefited from his technique (37).
outflow tract in previously atretic right ventricular outflow tract
Currently, two valves are being evaluated for the percutaneous
associated with ventricular septal defect. Am J Cardiol 1993; 72:
management of patients with severe conduit failure due to 354–356.
regurgitation and or stenosis. The Medtronic Melody valve 13. Hijazi ZM, Patel H, Cao QL, et al. Transcatheter retrograde radio-
developed by Bonhoeffer (Fig. 47.6) consists of a bovine jugular frequency perforation of the pulmonic valve in pulmonary atresia
vein with a valve inside sewn into a Platinum Cheatham stent. with intact ventricular septum, using a 2 French catheter. Cathet
The Edwards SAPIEN THV (38) is made of bovine pericardial Cardiovasc Diagn 1998; 45:151–154.
leaflets sewn inside a stainless steel stent. Both valves are 14. Humpl T, Soderberg B, McCrindle BW, et al. Percutaneous balloon
balloon expandable and require a large delivery system for valvotomy in pulmonary atresia with intact ventricular septum:
deployment (up to 22–24 Fr). Clinical trials in the United States impact on patient care. Circulation 2003; 108:826–832.
15. Allen HD, Beekman RH III, Garson A Jr., et al. Pediatric thera-
are under way to assess their safety and efficacy.
peutic cardiac catheterization: a statement for healthcare profes-
sionals from the Council on Cardiovascular Disease in the Young,
CONCLUSIONS American Heart Association. Circulation 1998; 97:609–625.
This chapter has summarized interventional techniques used 16. Gay BB Jr., French RH, Shuford WH, et al. The Roentgenologic
for treating CHDs in the catheterization laboratory. Adult CHD Features of Single and Multiple Coarctations of the Pulmonary
Artery and Branches. Am J Roentgenol Radium Ther Nucl Med
is a growing field and requires skilled personnel trained in both
1963; 90:599–613.
adult and CHDs who understand the physiology and anatomy
17. Bergersen L, Jenkins KJ, Gauvreau K, et al. Follow-up results of
of the cardiac lesion. The practice of interventional cardiology Cutting Balloon angioplasty used to relieve stenoses in small
for CHD also requires a fully equipped catheterization labora- pulmonary arteries. Cardiol Young 2005; 15:605–610.
tory to help minimize the risks of such procedures. Surgical 18. Fogelman R, Nykanen D, Smallhorn JF, et al. Endovascular stents
backup is crucial in any institution practicing congenital car- in the pulmonary circulation. Clinical impact on management and
diac intervention. Lastly, cardiac intervention is a rapidly medium-term follow-up. Circulation 1995; 92:881–885.
[22/6/010/11:53:2] [593–601]
19. O’Laughlin MP, Perry SB, Lock JE, et al. Use of endovascular 31. Sharma S, Kothari SS, Krishnakumar R, et al. Systemic-to-pulmo-
stents in congenital heart disease. Circulation 1991; 83:1923–1939. nary artery collateral vessels and surgical shunts in patients with
20. O’Laughlin MP, Slack MC, Grifka RG, et al. Implantation and cyanotic congenital heart disease: perioperative treatment by
intermediate-term follow-up of stents in congenital heart disease. transcatheter embolization. AJR Am J Roentgenol 1995;
Circulation 1993; 88:605–614. 164:1505–1510.
21. Rao PS, Galal O, Smith PA, et al. Five- to nine-year follow-up 32. Dutton JA, Jackson JE, Hughes JM, et al. Pulmonary arteriovenous
results of balloon angioplasty of native aortic coarctation in infants malformations: results of treatment with coil embolization in 53
and children. J Am Coll Cardiol 1996; 27:462–470. patients. AJR Am J Roentgenol 1995; 165:1119–1125.
22. Ing FF. Delivery of stents to target lesions: techniques of intra- 33. Lee DW, White RI Jr., Egglin TK, et al. Embolotherapy of large
operative stent implantation and intraoperative angiograms. pulmonary arteriovenous malformations: long-term results. Ann
Pediatr Cardiol 2005; 26:260–266. Thorac Surg 1997; 64:930–939; discussion 9–40.
23. McNamara JJ, Gross RE. Congenital coronary artery fistula. Sur- 34. Mager JJ, Overtoom TT, Blauw H, et al. Embolotherapy of pul-
gery 1969; 65:59–69. monary arteriovenous malformations: long-term results in 112
24. Alkhulaifi AM, Horner SM, Pugsley WB, et al. Coronary artery patients. J Vasc Interv Radiol 2004; 15:451–456.
fistulas presenting with bacterial endocarditis. Ann Thorac Surg 35. White RI Jr., Lynch-Nyhan A, Terry P, et al. Pulmonary arterio-
1995; 60:202–204. venous malformations: techniques and long-term outcome of
25. Skimming JW, Walls JT. Congenital coronary artery fistula suggesting embolotherapy. Radiology 1988; 169:663–669.
a “steal phenomenon” in a neonate. Pediatr Cardiol 1993; 14:174–175. 36. Bonhoeffer P, Boudjemline Y, Saliba Z, et al. Percutaneous replace-
26. Wilde P, Watt I. Congenital coronary artery fistulae: six new cases ment of pulmonary valve in a right-ventricle to pulmonary-artery
with a collective review. Clin Radiol 1980; 31:301–311. prosthetic conduit with valve dysfunction. Lancet 2000; 356:
27. Bauer HH, Allmendinger PD, Flaherty J, et al. Congenital coro- 1403–1405.
nary arteriovenous fistula: spontaneous rupture and cardiac tam- 37. Lurz P, Coats L, Khambadkone S, et al. Percutaneous pulmonary
ponade. Ann Thorac Surg 1996; 62:1521–1523. valve implantation: impact of evolving technology and learning
28. Ramo OJ, Totterman KJ, Harjula AL. Thrombosed coronary artery curve on clinical outcome. Circulation 2008; 117:1964–1972.
fistula as a cause of paroxysmal atrial fibrillation and ventricular 38. Garay F, Webb J, Hijazi ZM. Percutaneous replacement of pulmo-
arrhythmia. Cardiovasc Surg 1994; 2:720–722. nary valve using the Edwards-Cribier percutaneous heart valve:
29. Qureshi SA. Coronary arterial fistulas. Orphanet J Rare Dis 2006; 1:51. first report in a human patient. Catheter Cardiovasc Interv 2006; 67:
30. Kharouf R, Cao QL, Hijazi ZM. Transcatheter closure of coronary 659–662.
artery fistula complicated by myocardial infarction. J Invasive
Cardiol 2007; 19:E146–E149.
[2/7/010/10:14:50] [602–609]
48
Embolization for fistulas and AVMs

Nicholas Kipshidze, Robert Rosen, and Irakli Gogorishvili
PERIPHERAL ARTERIOVENOUS middle sacral artery, lumbar artery, and femoral arteries (3).
MALFORMATIONS Extremity AVM (Fig. 48.2) symptoms range from mild swell-
Introduction ing or soft tissue mass, extremity overgrowth or growth
An arteriovenous fistula is an abnormal connection between an retardation, bleeding, ulceration, gangrene, or rarely conges-
artery and a vein. As a consequence, the blood bypasses the tive heart failure.
capillary bed. The majority of arteriovenous malformations
(AVMs) are congenital, but some can be acquired due to trauma,
infection, or malignancy. An arteriovenous fistula may be the
Indications for Treatment
Treatment of AVMs, especially high-flow AVMs, is challenging
result of a vascular catheter insertion or may be created surgi-
and not always successful regardless of the type of procedure
cally to provide access for hemodialysis in end-stage kidney
chosen (percutaneous, surgical, or combined). Treatment
failure patients. AVMs can occur anywhere in the body, includ-
should therefore be undertaken only when clinically indicated
ing the central nervous system. Congenital AVMs are present at
(3). Asymptomatic lesions that are discovered incidentally gen-
birth and, although they may be asymptomatic, they always
erally do not require treatment. Absolute and relative indica-
persist and do not involute (1,2), unlike true hemangiomas.
tions for treatment are as follows:
AVMs may become symptomatic with age, sometimes at puberty
Absolute indications for treatment
in female patients, during pregnancy, or after local trauma.
Of special interest are pulmonary AVMs, mostly related to l Hemorrhage, major or recurrent minor
hereditary hemorrhagic telangiectasia (Osler–Weber–Rendu l Gangrene or ulcer of arterial, venous, or combined origin
syndrome), which is discussed later in this chapter. l Ischemic complication of acute and/or chronic arterial
insufficiency
Anatomic Considerations l Progressive venous complication of chronic venous insuf-
Vascular malformations are divided into low- and high-flow ficiency with venous hypertension
l High-output cardiac failure (clinical and/or laboratory)
lesions. Low-flow malformations are venous, lymphatic, or
l Lesion located at life-threatening vital areas that compro-
mixed—therefore, causing congestion of either type. They
will be discussed later. High-flow lesions contain an arterial mise vision, hearing, eating, or breathing
component and determine left-to-right shunt. Relative indications for treatment
These subtypes differ from each other with respect to
l Various symptoms and signs affecting the quality of life;
clinical course and treatment options. High-flow malformations
are less common, more difficult to treat, and are prone to disabling pain and/or functional impairment
l Lesions with a potentially high risk of complications
recurrence. High-flow AVMs result in a similar pathophysiol-
ogy to arteriovenous fistulas and may result in: (e.g., hemarthrosis) and/or limb-threatening location
l Lesions causing limb length discrepancy
1. Compressive and erosive effects l Cosmetically severe deformity with or without functional
2. Venous stasis disability
3. Ischemia due to peripheral steal phenomenon
4. High-output heart failure
Principles of Treatment
Although AVMs can occur anywhere in the body, the The mainstay of treatment of high-flow AVMs is permanently
pelvis and the extremities are the most common locations for closing or eliminating the vascular nidus where arterial blood is
peripheral AVMs. shunted to the veins.
Historically, surgical treatment alone proved to be inad-
Clinical Aspects equate or even disastrous, often leading to extensive damage to
Pelvic AVMs (Fig. 48.1) may produce pain, pelvic venous adjacent structures with high recurrence rates or major ampu-
congestion, sexual dysfunction, and, occasionally, high-output tation (4,5). The reason for failure is that it is rarely possible to
cardiac failure and hemorrhage. With respect to hemorrhage, completely resect the nidus, which can be large or supplied by
this is generally a lower gastrointestinal bleed or hematuria, multiple collaterals. Proximal ligation of feeding arterial
while hemorrhage in the abdominal cavity is extremely rare. branches is ineffective as aggressive recruitment of new feeding
Although the most frequent blood supply of pelvic AVMs collaterals is common (6,7).
arises from the hypogastric arteries, there may also be multi- Disappointing results of surgical treatment stimulated
ple feeding branches from the inferior mesenteric artery, development of alternative approaches. The introduction of
[2/7/010/10:14:50] [602–609]
EMBOLIZATION FOR FISTULAS AND AVMs 603
Figure 48.3 Schematic drawing of AVM “nidus,” showing that only

complete occlusion of all feeding branches will result in complete
AVM elimination. Abbreviation: AVM, arteriovenous malformation.
Figure 48.1 Pelvic AVM. Abbreviation: AVM, arteriovenous mal- embolotherapy in the early 1970s (8,9) and advancements in
formation. catheter technology made it possible to treat AVMs less inva-
sively. Although AVMs are uncommon, data about interven-
tions and follow-up is increasing (10–15). Some centers
document very promising results with interventional treat-
ment. However, several reports have documented worsening
of symptoms after embolization (16). This is why most authors
recommend intervention only for patients with significant
symptoms (11,13).
Embolization Technique
The goal of treatment is to occlude the nidus of the malforma-
tion through selective catheterization and embolization of feed-
ing branches (Fig. 48.3).
Embolization of the nidus of an AVM often requires
super-selective catheterization of numerous arterial feeding
branches. This is facilitated by use of coaxial microcatheter
systems. A 2- to 3-Fr microcatheter is coaxially introduced
through a 4- to 5-Fr selective catheter and can be manipulated
into the terminal feeding artery. Embolic agents are then
delivered via the microcatheter, an ideal tool for the delivery
of liquid agents, particles, and small coils.
Embolic Agents for High-Flow Lesions

Use of the proper embolic agent (Table 48.1) is critical when
treating AVMs, as the wrong agent may not only fail to treat
the lesion but may also interfere with future attempts at
treatment. In most cases, the goal is penetration and eradi-
cation of the nidus of the lesion. Fibered coils (Fig. 48.4) have
been the material of choice for vessel occlusion since their
introduction 30 years ago (17). However, their value in treat-
ing AVMs is limited as they result in proximal occlusion,
which is functionally equivalent to surgical ligation. There-
Figure 48.2 Extremity AVM. A 23-year-old soldier with traumatic fore, fibered coils are appropriate only in lesions with a
arteriovenous fistula treated later with stent graft. Source: Courtesy fistula-like architecture.
of Kipshidze University Hospital. Abbreviation: AVM, arteriovenous The basic fibered coil consists of a length of guidewire
malformation. with multiple polyester threads attached transversely along
most of its length. Fibered coil emboli are preshaped into a
[2/7/010/10:14:50] [602–609]
Table 48.1 Embolic Agents Detachable balloons were used to occlude large arterio-
venous communications, but their complexity and cost remain
l Metallic coils
l
significant drawbacks. In addition, they are not commercially
Detachable balloons
l
available in the United States at this time.
Covered stents
l Particulate agents Polyvinyl alcohol particles ranging from <100 mm to
l Polyvinyl alcohol particles >1000 mm in size are used for permanent vessel embolization.
l Trisacryl gelatin microspheres (Embosphere) The particles wedge in vessels of corresponding diameter and
l Ytrium-90 glass radioactive microspheres (TheraSphere) produce a permanent occlusion by thrombosis and subsequent
l Embogold fibrosis. The particle-contrast suspension is injected slowly in
l Gelfoam small aliquots under continuous fluoroscopic control once the
l N-butyl 2-cyanoacrylate delivery catheter is positioned in the desired location. While the
l Onyx particles themselves are permanent, the effect on the malfor-
l Liquid sclerosing agents
l
mation is often temporary, with collateral recruitment and
Ethanol or absolute alcohol
l
recanalization around the particles.
Hypertonic dextrose
l Sotradecol Liquid occlusive agents, including sclerosants and glue,
l Ethibloc are very attractive for AVM nidus occlusion (Fig. 48.5). Absolute
ethanol is a very potent sclerosant and should be used with great
care. Although good long-term results have been reported, intra-
arterial ethanol injection carries a significant risk of damage to
normal tissues if nontarget embolization occurs.
Glue or tissue adhesives, such as N-butyl cyanoacrylate
and isobutyl cyanoacrylate, are another category of liquid
embolic agents that polymerize on contact with an ionic envi-
ronment such as blood. In experienced hands, these are very
effective for AVM occlusion, but require frequent, time-con-
suming catheter exchanges, and as with ethanol, great care
should be taken to avoid nontarget embolization.
Complications
Embolotherapy, although minimally invasive and relatively
safe, is not without complications. Therefore, patients need to
be fully informed of the potential risks of the treatment. Prob-
ably the most common complication is ischemia due to non-
target embolization or distal migration of the embolic agent.
Sometimes superficial tissue necrosis occurs, and extensive
lesions may require skin grafting.
Postembolization syndrome caused by tissue necrosis
may be encountered after embolotherapy. The symptoms of
pain, fever, leukocytosis, and nausea arise shortly after embo-
lotherapy and usually resolve within a few days; however, they
may persist up to a week. Compared to tumor embolization
and organ embolotherapy, postembolization syndrome occurs
Figure 48.4 Nester coils (Cook Inc., Bloomington, Indiana, U.S.). less frequently after the treatment of AVMs.
PULMONARY ARTERIOVENOUS
MALFORMATIONS
variety of different configurations and then stretched out in a Pulmonary arteriovenous malformations (PAVMs) are discussed
cartridge for delivery into a catheter. Coils are made of steel separately because of their unique pathologic mechanisms, asso-
and platinum with spring sizes 0.035 to 0.038 and 0.018 in. ciated risk factors, and different treatment approaches. Like
Selecting the correct coil size is extremely important for the other arteriovenous fistula malformations, PAVMs consist of a
success of the procedure. The coils should be slightly oversized congenital connection between a pulmonary artery and vein
relative to the diameter of the target vessel to allow them to without normal capillary bed. Unlike other AVMs, the shunt is
grip the vessel wall and be closely packed. Significantly over- from right to left. This carries significant implications in terms
sized coils tend to pass through the vessel like a guidewire and of clinical presentation and treatment. Approximately 70% of
may elongate to a feeding branch rather than coiling up. PAVMs are congenital and multiple (19,20) and 60% to 90% of
Undersized coils may fail to lodge and migrate causing unin- them are associated with hereditary hemorrhagic telangiectasia,
tended embolization into other vessels. also called Osler–Weber–Rendu syndrome. The remainder are
Guglielmi detachable coils without fibers were intro- isolated lesions in otherwise healthy patients with no family
duced for controlled and safer embolization. But their high history. Rarely, PAVMs are acquired and related to cirrhosis,
cost and limited thrombogenicity confine their use to treatment congenital heart disease (Glenn and Fontan shunts), tumors,
of intracerebral aneurysms. trauma, and infection.
[2/7/010/10:14:50] [602–609]
Figure 48.5 Arteriovenous malformation involving the right leg of a teenage boy. (A–D) Preintervention angiography. The steps for primary
embolization using permanent liquid agents (E,F—direct injection) and placing adjunctive coils (G,H) are shown. The patient had an excellent
result after four embolizations carried out in two hospital visits, and he is now symptom-free. Source: From Ref. 18.
Anatomic Considerations and Clinical Aspects orthodeoxia may be seen), chest X ray, contrast echocardiog-
PAVMs are classified as simple (80–90% of cases), which are raphy (high sensitivity >90%), computed tomographic (CT)
supplied by an artery contained within one pulmonary seg- angiography, magnetic resonance imaging (MRI), and finally
ment, or complex, which are fed by arteries from more than one pulmonary angiography—considered the gold standard for
pulmonary segment (Figs. 48.6 and 48.7). Approximately 65% diagnosis. In recent years, the availability of multidetector CT
of the lesions are located in lower lobes (22). PAVMs vary in scanning has allowed noninvasive diagnosis and reconstruc-
size from tiny, angiographically invisible structures to giant tion of pathologic anatomy with high resolution (25).
malformations occupying an entire lobe (Fig. 48.8).
Up to 55% of PAVMs are asymptomatic and are discov-
ered incidentally. Those that are symptomatic can present in a Treatment Considerations
variety of ways (23). The most common presentations are The initial treatment for PAVMs was surgical including vascu-
dyspnea, cyanosis, easy fatigability, clubbing, hemoptysis, or lar ligation, lobectomy, or pneumonectomy and was associated
hemothorax. Pathologic mechanisms related to PAVM are with up to10% recurrence and serious morbidity. In 1977,
hypoxemia due to right-to-left shunt, rupture and bleeding of Porstmann performed the first PAVM embolization and over
pathologic vessels, and paradoxical embolism due to loss of time this has become the standard of treatment.
lung-filtering function. Paradoxical embolism is the main com- The embolization procedure is generally performed from
plication associated with PAVMs. It is estimated that in patients a femoral approach using 6- to 7-Fr guiding catheters posi-
with PAVMs the lifetime probability of transient ischemic tioned in the feeding artery. Embolization is performed using
attack or stroke is 25% and of cerebral abscess is 10% (23,24). one of a variety of macroscopic devices, including coils, micro-
For this reason, PAVMs tend to be treated more aggressively coils, detachable plugs, and detachable balloons. During the
than other lesions, even when asymptomatic. procedure, the patient is anticoagulated, commonly with 5000
units of heparin, and meticulous care must be taken to avoid air
bubble injection, which can result in procedure-related stroke.
Indications for PAVM Treatment The most popular embolization devices for PAVMs today
A PAVM with >3 mm feeding artery is generally considered an are coils. There are several techniques of deploying different
indication for intervention. Diagnostic tests for pulmonary types of coils (Figs. 48.9–48.11). Usually the first coil selected
vascular malformation include pulse oximetry (platypnea has a diameter at least 20% larger than the vessel to be
[2/7/010/10:14:50] [602–609]
Figure 48.8 Angiography in left anterior oblique projection dem-

onstrating a huge PAVM in the right lung of a 24-year-old woman.
Abbreviation: PAVM, pulmonary arteriovenous malformation.
Figure 48.6 Diagrammatic examples of simple PAVMs (supplied

with only one feeding artery). The nidus on the lower two panels
consists of a network of small branches and septations. Abbrevia-
tions: PAVM, pulmonary arteriovenous malformation; PA, pulmo-
nary artery; PV, pulmonary vein. Source: From Ref. 21.
Figure 48.9 Anchor technique. The guide catheter maintains posi-

Figure 48.7 Complex pulmonary arteriovenous malformation. tion in the artery and the first 1 to 2 cm of coil is placed in the side
Abbreviations: PA, pulmonary artery; PV, pulmonary vein. Source: branch. The remaining coil is deposited as a tight coil mass. Source:
From Ref. 21. Courtesy of Cook, Inc.
[2/7/010/10:14:50] [602–609]
Figure 48.10 Scaffold technique. Deployment of first higher radial

force coil began the formation of a “nidus” or “endoskeleton” for
placement of additional coils. Source: Courtesy of Cook, Inc. Figure 48.12 Amplatzer vascular plug (AGA Medical, Golden Val-
ley, Minnesota, U.S.).
(Fig. 48.12). Figure 48.13 shows two examples of complex

PAVMs treated with coils.
Procedural success in recent studies is 85% to 95% (26,27).
Although overall results during follow-up have been satisfac-
tory, in patients with diffuse PAVMs, improvement of dyspnea,
oxygenation, and shunt fraction may not be complete (28).
Major neurologic complications such as cerebral abscess, tran-
sient ischemic attack, or stroke related to reperfused or newly
perfused PAVMs also have been reported (27). On the basis of
this experience, helical CT scans at 6 and 12 months and then
every 3 to 5 years afterward are recommended. Complications
of the procedure are uncommon and include device migration
(1%) and air embolism (up to 4%). The most frequent symptom
postprocedure is pleuritic pain (13%), which usually resolves
Figure 48.11 Coaxial technique to prevent coil elongation. To within 24 hours.
achieve permanent occlusion with fibered coils, it is useful to have
a guide catheter, proximal in the artery, and use a smaller coaxial
catheter to tightly pack (nest) the coils and achieve cross-sectional LOW-FLOW VASCULAR MALFORMATIONS
occlusion. Source: Courtesy of Cook, Inc. Low-flow malformations include venous and lymphatic
lesions. Venous malformations are the most common type of
congenital lesion encountered clinically. They may occur any-
where in the body but have a predilection for the lower
extremities. Venous malformations may be of the cavernous
occluded. Cross-sectional occlusion is the goal to minimize the type (sometimes incorrectly referred to as cavernous heman-
chance of future recanalization. It is important to perform giomas) or consist of abnormal venous channels. In some
the embolization with coils placed as distally as possible in the patients, such as those with Klippel–Trenaunay syndrome,
feeding vessel. This technique avoids the occlusion of branches there may be a combination of both types of lesion.
to normal lung and reduces the risk of pleuritic pain or pul- Presenting symptoms may include pain, swelling, cuta-
monary infarction. Detachable balloons have the advantage of neous lesions causing bleeding, venous hypertension, and
flow guidance, but have largely been replaced by detachable spontaneous thrombosis. Some of the venous syndromes may
coils and plugs that are easier to use. Amplatzer vascular plugs, be associated orthopedic or growth disturbances. Evaluation is
which are made of tightly woven nitinol mesh, can be reposi- by ultrasound, CT, and MRI, where they are particularly well
tioned prior to detachment and have provided excellent results demonstrated.
[2/7/010/10:14:50] [602–609]
Figure 48.13 Two complex PAVMs (A) successfully treated with coils (B). Abbreviation: PAVMs, pulmonary arteriovenous malformations.
Conservative treatment may be effective in mild cases, REFERENCES

with measures such as elevation and elastic support. For larger 1. Belov S. Anatomopathological classification of congenital vascular
or more symptomatic lesions, embolization techniques can be defects. Semin Vasc Surg 1993; 6:219–224.
highly effective. Since these are by definition venous lesions, 2. Mulliken JB, Glowacki J. Classification of pediatric vascular
embolizing the arterial supply to the area is generally ineffec- lesions. Plast Reconstr Surg 1982; 70:120–121.
3. Zhou W, Lin PH, Eraso A, et al. Embolotherapy of peripheral
tive. Direct injection of sclerosing agents is the mainstay of
arteriovenous malformations. Endovasc Today 2005:85–91.
treatment, using agents such as ethanol and sodium tetradecyl 4. Coursley G, Ivins JC, Barker NW. Congenital arteriovenous fistu-
sulfate (Sotradecol). These agents cause thrombosis of the las in the extremities; an analysis of sixty-nine cases. Angiology
venous spaces, followed by an inflammatory reaction, and 1956; 7:201–217.
finally shrinkage and fibrosis. While multiple treatments may 5. Gomes MM, Bernatz PE. Arteriovenous fistulas: a review and
be required, clinical improvement or resolution of symptoms ten-year experience at the Mayo Clinic. Mayo Clin Proc 1970;
can be achieved in most patients. 45:81–102.
Lymphatic malformations are less common and more 6. Malan E. Surgical problems in the treatment of congenital arterio-
difficult to treat. They range from large cystic lesions (cystic venous fistulae. J Cardiovasc Surg (Torino) 1965; (suppl):251–255.
hygroma) to cutaneous lesions with vesicle formation. One of 7. Tille JC, Pepper MS. Hereditary vascular anomalies: new insights
into their pathogenesis. Arterioscler Thromb Vasc Biol 2004;
the distinguishing characteristics of lymphatic malformations is
24:1578–1590.
their tendency to infection. The presence of infection in a 8. Grace DM, Pitt DF, Gold RE. Vascular embolization and occlusion
vascular malformation nearly always indicates presence of a by angiographic techniques as an aid or alternative to operation.
lymphatic component. Surg Gynecol Obstet 1976; 143:469–482.
Treatment varies according to the type of lesion. Large 9. Yazaki T, Tomita M, Akimoto M, et al. Congenital renal arterio-
cystic lesions can be treated surgically or by percutaneous venous fistula: case report, review of Japanese literature and
drainage followed by sclerotherapy. Agents shown to be par- description of non-radical treatment. J Urol 1976; 116:415–418.
ticularly effective in lymphatic lesions include OK-432, bleo- 10. Jacobowitz GR, Rosen RJ, Rockman CB, et al. Transcatheter
mycin, and doxycycline. Superficial lesions are particularly embolization of complex pelvic vascular malformations: results
difficult to manage and may require surgical resection and/or and long-term follow-up. J Vasc Surg 2001; 33:51–55.
11. Rockman CB, Rosen RJ, Jacobowitz GR, et al. Transcatheter
laser therapy.
embolization of extremity vascular malformations: the long-term
success of multiple interventions. Ann Vasc Surg 2003; 17:417–423.
CONCLUSIONS 12. Sofocleous CT, Rosen RJ, Raskin K, et al. Congenital vascular
malformations in the hand and forearm. J Endovasc Ther 2001;
Over the past two decades, catheter-based embolization has
8:484–494.
assumed a primary role in the management of AVMs. With 13. Tan KT, Simons ME, Rajan DK, et al. Peripheral high-flow arterio-
improvements in technology, procedures have become increas- venous vascular malformations: a single-center experience. J Vasc
ingly safe and effective. While peripheral AVMs are usually Interv Radiol 2004; 15:1071–1080.
treated only in the presence of symptoms, PAVMs may be 14. Upton J, Coombs CJ, Mulliken JB, et al. Vascular malformations of
treated in asymptomatic patients to prevent ischemic neuro- the upper limb: a review of 270 patients. J Hand Surg Am 1999;
logic events or cerebral abscesses. 24:1019–1035.
[2/7/010/10:14:50] [602–609]
15. White RI Jr., Pollak J, Persing J, et al. Long-term outcome of 22. White RI Jr., Mitchell SE, Barth KH, et al. Angioarchitecture of
embolotherapy and surgery for high-flow extremity arteriovenous pulmonary arteriovenous malformations: an important consider-
malformations. J Vasc Interv Radiol 2000; 11:1285–1295. ation before embolotherapy. AJR Am J Roentgenol 1983; 140:681–686.
16. Dickey KW, Pollak JS, Meier GH III, et al. Management of large 23. Shovlin CL, Guttmacher AE, Buscarini E, et al. Diagnostic criteria
high-flow arteriovenous malformations of the shoulder and upper for hereditary hemorrhagic telangiectasia (Rendu-Osler-Weber
extremity with transcatheter embolotherapy. J Vasc Interv Radiol syndrome). Am J Med Genet 2000; 91:66–67.
1995; 6:765–773. 24. White RI Jr. Recanalization after embolotherapy of pulmonary
17. Gianturco C, Anderson JH, Wallace S. Mechanical devices for arteriovenous malformations: significance? Outcome? AJR Am J
arterial occlusion. Am J Roentgenol Radium Ther Nucl Med 1975; Roentgenol 1998; 171:1704–1705.
124:428–435. 25. Remy J, Remy-Jardin M, Wattinne L, et al. Pulmonary arterio-
18. Rutherford RB, ed. Vascular Surgery. 6th ed. Philadelphia, PA: venous malformations: evaluation with CT of the chest before and
Saunders, 2005. after treatment. Radiology 1992; 182:809–816.
19. Gossage JR, Kanj G. Pulmonary arteriovenous malformations. A 26. Mager JJ, Overtoom TT, Blauw H, et al. Embolotherapy of pulmo-
state of the art review. Am J Respir Crit Care Med 1998; 158:643–661. nary arteriovenous malformations: long-term results in 112 patients.
20. Shovlin CL, Letarte M. Hereditary haemorrhagic telangiectasia J Vasc Interv Radiol 2004; 15:451–456.
and pulmonary arteriovenous malformations: issues in clinical 27. Remy-Jardin M, Wattinne L, Remy J. Transcatheter occlusion of
management and review of pathogenic mechanisms. Thorax 1999; pulmonary arterial circulation and collateral supply: failures,
54:714–729. incidents, and complications. Radiology 1991; 180:699–705.
21. White RI Jr., Pollak JS, Wirth JA. Pulmonary arteriovenous mal- 28. Faughnan ME, Lui YW, Wirth JA, et al. Diffuse pulmonary
formations: diagnosis and transcatheter embolotherapy. J Vasc arteriovenous malformations: characteristics and prognosis.
Interv Radiol 1996; 7:787–804. Chest 2000; 117:31–38.
[23/6/010/11:13:38] [610–624]
49
Lower and upper extremity intervention

Tahir Mohamed and Mark Robbins
LOWER EXTREMITY and popliteal arteries constitute the outflow vessels while the
Introduction anterior tibial (AT), posterior tibial (PT), and peroneal arteries
Atherosclerotic peripheral arterial disease (PAD) is highly are commonly known as runoff vessels. Different locations
prevalent, present in at least 8 to 12 million adults living in within the lower extremity vasculature deserve special attention
the United States. As the population ages, its incidence will when performing endovascular procedures as not all areas carry
undoubtedly continue to increase (1). PAD can have a major the same procedural risk and some may be more device specific
impact on quality of life, and it remains a significant source of as generalized in Table 49.1. The proximal CFA and the iliac
morbidity and mortality. Because atherosclerosis is a systemic arteries are retroperitoneal and therefore are not constrained by
process, patients with PAD are at high risk of coronary and soft tissue as the SFA. Therefore, perforations within the prox-
cerebrovascular ischemic events, and thus the diagnosis of PAD imal CFA and iliac arteries can lead to rapid blood loss and
has strong prognostic implications. Over the last 10 years, there should be considered an emergency. As will be discussed later,
has been an exponential growth in the number of endovascular emergency equipment must be available and ready to treat this
repair procedures for lower extremity claudication and critical life-threatening complication. The CFA should not be stented
limb ischemia. The procedures have grown in almost every because it is the site for vascular entry and a flexure point.
subspecialty including cardiology, vascular surgery, and inter- Similarly, stenting the popliteal artery should be avoided if
ventional radiology. Advances in technology have allowed the possible. Both the CFA and the popliteal artery are ideal
development of more effective endovascular approaches that locations for rotational or directional atherectomy as well as
has prompted a swing away from open surgical repair. This novel angioplasty techniques. The distal SFA—as it enters into
being accepted we must continue to recognize which modality the adductor hiatus (Fig. 49.2)—is a common location for occlu-
is the best suited to assure the lowest risk procedure that result sive vascular disease. This location is also associated with an
in the most durable outcome. increased incidence of stent fractures due to torsion during knee
flexion and extension. Debulking the distal SFA with or without
adjunctive angioplasty remains a valuable option for this loca-
Indications and Evidence Based Approach tion to limit stent fracture and associated restenosis. Below the
The decision to perform a revascularization procedure—surgical knee repair has historically been withheld for severe limb
or endovascular—in a patient with PAD cannot be made exclu- ischemia and primarily treated with open surgical repair. How-
sively on the basis of the angiographic extent of disease. Patients ever, endovascular repair continues to gain acceptance for
should be selected on the basis of the severity of the symptoms, infrapopliteal disease as will be discussed later in the chapter.
disability as assessed by the patient and physician, failure of
medical therapy, and a favorable risk/benefit ratio. In patients
Fundamentals
with intermittent claudication (IC) limiting quality of life and in
The basic fundamentals for performing endovascular repair once
those who progress to ischemic rest pain, ulceration, or gangrene
indications have been satisfied begin with an in-depth under-
revascularization attempts are warranted. Guidelines for the
standing of vascular anatomy, collateral supply, and the ability to
performance of endovascular treatment for claudication have
perform an imaging work-up, allowing for a detailed planning of
been published by a joint ACC/AHA task force (2). In general,
the upcoming intervention. There are pathways to gain clinical
endovascular repair is most successful when utilized in the
competence either through a dedicated training program or phy-
larger inflow arteries with discrete lesions. Less favorable long-
sician experience tract (2,3). Comprehensive knowledge of the
term outcomes are associated with endovascular repair of dif-
equipment available is another vital component of a successful
fuse disease in smaller outflow and runoff vessels. Comorbid
endovascular program. Unlike coronary equipment, peripheral
conditions also play a role in the decision process with more
interventional devices may be 0.014, 0.018, or 0.035 in. guidewire
aggressive endovascular approaches being accepted in patients
compatible and may require varying sheath and guide diameters.
at higher risk for open surgical repair.
In addition, the working length of devices may range from 70 to
150 cm. The main focus of this chapter will be on the fundamental
Anatomic Considerations technical aspects of endovascular repair.
The lower extremity arterial supply begins with the bifurcation
of the common iliacs from the distal aorta. It is then subdivided Equipment
into inflow, outflow, and runoff vessels as outlined in Fig- Sheaths
ure 49.1. Inflow arteries include the common, external, and The sheath becomes the delivery catheter for most endovascu-
internal iliacs as well as a portion of the common femoral artery lar repair procedures. The main advantage to using sheaths is
(CFA). The superficial femoral artery (SFA), profunda femoris, that its internal diameter is much larger than the one of a same
[23/6/010/11:13:38] [610–624]
LOWER AND UPPER EXTREMITY INTERVENTION 611
Figure 49.2 Adductor hiatus (Hunter’s canal).
aortic interventions can be performed via a 6-Fr sheath but

occasionally a 7-Fr sheath will be needed for larger stent and
post dilatation balloons. CFA and SFA procedures typically call
for crossing over the aortic bifurcation from the contralateral
CFA, which requires a sheath at least 45 cm in length. A 6-Fr
Figure 49.1 Major arteries of the lower extremities. diameter is usually adequate to deliver most of the interven-
tional equipment (see exceptions listed in Tables 49.3–49.6).
Although most infrapopliteal interventions can be performed
with 45-cm 6-Fr sheaths on occasion a 65-cm length sheath
placed deep into the SFA may provide better visualization of
the target vessel. When approaching the lower extremity via
size guide catheter. Length and diameter of the sheath to be the brachial artery a 90-cm 6- or 7-Fr sheath is usually adequate
used will depend on distance from access site to lesion and the for iliac and CFA lesions.
internal diameter required to deliver the interventional equip-
ment. Most iliac and distal aortic interventions will be per- Wires
formed via unilateral or bilateral retrograde CFA access using Wires most commonly used for lower and upper extremity
sheaths that are at least 23 cm in length enabling adequate interventional procedures are listed in Table 49.2. When cross-
visualization of the lesion as seen in Figure 49.3. Most iliac and ing a nonocclusive stenosis, virtually any wire can be used for
Table 49.1 Lesion Characteristic and Optional Equipment for Endovascular Intervention
Lesion characteristic Device
Iliac and distal aorta PTA with adjunctive stent
CFA noncalcified SilverHawk, POBA, cryoplasty, cutting balloon
CFA calcified RockHawk, Diamondback 2.25 PTA
SFA ostial SilverHawk, POBA, cryoplasty, cutting balloon,
RockHawk, Diamondback 2.0–2.25 PTA for heavy calcification
SFA proximal and mid nonocclusive PTA, PTA þ self-expanding stent, cryoplasty, cutting balloon, SilverHawk, ELA
<5 cm in length RockHawk, Diamondback 2.25 PTA for heavy calcification
SFA proximal and mid nonocclusive PTA þ self-expanding stent, SilverHawk, ELA
>10 cm in length RockHawk, Diamondback 2.0–2.25 PTA for heavy calcification
SFA occlusion Subintimal angioplasty reentry catheter with adjunctive self-expanding stent ViabHan, ELA
SFA distal and popliteal SilverHawk, POBA, cryoplasty, cutting balloon, ELA
RockHawk, Diamondback 2.0–2.25 PTA for heavy calcification
Runoff vessels SilverHawk, POBA, cryoplasty, cutting balloon
Diamondback 1. 5–1.75 PTA for heavy calcification
Abbreviations: CFA, common femoral artery; ELA, excimer laser angioplasty; POBA, plain old balloon angioplasty; PTA, percutaneous
transluminal angioplasty; SFA, superficial femoral artery.
[23/6/010/11:13:38] [610–624]
Figure 49.3 (A) Stent in the external iliac artery extending into the distal aorta (arrowhead ) and 23-cm, 6-Fr CFA sheath extending into the
external iliac artery (arrow ); (B) final result after stent placement. Abbreviation: CFA, common femoral artery.
Table 49.2 Examples of Guidewires Used for Endovascular Table 49.3 Selected Examples of Angioplasty Balloons for
Intervention Endovascular Intervention
Guidewire Indication Device Minimal
Guidewire 0.035 in. CTO, tortuous calcified vessels diameter sheath
(Terumo Medical) Device (mm) (Fr) Wire RX/OTW
Magic Torque 0.035 in. Nonocclusive stenosis, Aviator (Cordis) 4–6.5 4 0.014 RX
(Boston Scientific) good support 7 5 0.014 RX
Supracore 0.035 in. Nonocclusive stenosis, ViaTrac (Abbott) 4–4.5 4 0.014 RX
(Abbott Vascular) good support 5–7 5 0.014 RX
Wholey 0.035 in. Nonocclusive stenosis, Sterling (Boston 3–8 4 0.014 RX
(Mallinckrodt) good support Scientific)
Amplatz Super Stiff 0.035 in. Extra support wire for railing Savvy (Cordis) 2–6 5 0.018 OTW
(Boston Scientific) purposes AgilTrac (Abbott) 4–8 5 0.018 OTW
Steelcore 0.018 in. Smaller profile balloons 9–10 6 0.018 OTW
(Abbott Vascular) 12–14 7 0.035 OTW
Confianza Pro 0.014 in. Penetrating distal cap of CTO Powerflex (Cordis) 4–8 6 0.035 OTW
(Abbott Vascular, Asahi) 9–10 7 0.035 OTW
Abbreviation: CTO, chronic total occluded vessel. Dorado (Bard) 3–7 5 0.035 OTW
7–10 6 0.035 OTW
EverCross (ev3) 3–6 5 0.035 OTW
7–10 6 0.035 OTW
advancement of balloons and stents as long as it provides 12 7 0.035 OTW
adequate support. Therefore, 0.035 in. and 0.018 in. wires are Abbreviations: RX, rapid exchange; OTW, over the wire.
usually chosen to deliver larger devices. Chronic total occlu-
sions (CTOs) more commonly require a hydrophilic-coated Table 49.4 Predictors of Poor Outcome After Plain Old Balloon
wire that will penetrate the initial cap and allow for subintimal Angioplasty
dissection and distal reentry to the true lumen. Certain devices
l Length of the diseased segment
require a 0.014 in. platform for their delivery (Tables 49.3–49.6).
l Total occlusion
On occasion, reentering the true lumen with the tapered
l Diabetes mellitus
Confianza Pro 0.014 in. wire (Abbott, Abbott Park, Illinois, l Poor distal runoff
U.S.) can be accomplished when unsuccessful with the hydro- l Critical limb ischemia
philic 0.035 in. wires and prior to using reentry devices.
Angioplasty Balloons availability of longer balloons for peripheral arterial use,

A select sample of balloons used for lower extremity angio- plain old balloon angioplasty (POBA) remains limited by resid-
plasty is listed in Table 49.3. Despite improvements in balloon ual dissection, acute elastic recoil, and restenosis (4). Although
catheter materials, lower balloon crossing profiles, and the POBA can be an effective modality for focal lesions in the iliac
[23/6/010/11:13:38] [610–624]
Table 49.5 Selected Examples of Balloon and Self-Expandable Stents for Endovascular Intervention
Device diameter (mm) Minimal sheath (Fr) Wire RX/OTW
Balloon expandable
Palmaz Genesis (Cordis) 6–7 6 0.035 OTW
8–10 7 0.035 OTW
Express LD (Boston Scientific) 5–7 6 0.035 OTW
8–10 7 0.035 OTW
Self-expandable:
Smart Control (Cordis) 6–10 6 0.035 OTW
Protégé Everflex (ev3) 6–8 6 0.035 OTW
Absolute Pro (Abbott) 6–10 7 0.035 OTW
Supera (IDev Tech) 4–9 7 0.035 OTW
Sentinol (Boston Scientific) 5–10 6 0.035 OTW
Abbreviations: RX, rapid exchange; OTW, over the wire.
Table 49.6 Atherectomy Devices for Endovascular Repair

Device Catheter Minimal sheath (Fr) Wire RX/OTW
SilverHawk (ev3) DS 6 0.014 RX
SS, SX 7 0.014 RX
MS, LS, LX 8 0.014 RX
Diamondback (CSI) 1.5, 1.75, 2.0 6 Viper Wire OTW
2.25 7 Viper Wire OTW
Abbreviations: RX, rapid exchange; OTW, over the wire.
arteries (4), the results of balloon angioplasty for complex a need for bailout stenting resulting from a suboptimal angio-
infrainguinal arterial disease have been disappointing (5–8). graphic result in 9% of cases and flow-limiting dissection in 7%
Numerous clinical and lesion specific factors have been iden- of cases. Primary clinical patency was 83% at nine months.
tified that negatively affect the long-term results of POBA Primary assisted and secondary patency rates were 94% and
(Table 49.4). Locations where POBA is desirable are distal 98%, respectively (11). These findings compare favorably with
SFA, popliteal, CFA, and runoff vessels. Stent fracture makes POBA historical data.
stenting less desirable for distal SFA and popliteal arteries,
while reaccess makes CFA stenting a relative contraindication. Cutting and Scoring Balloons
Advances in technology for balloon angioplasty have included There is extensive experience with cutting and, to less extent,
the production of cutting and scoring balloons, as well as with the scoring balloons in the coronary circulation. Recently,
cryoplasty balloons—as described below—to help alleviate the cutting balloon (Boston Scientific)—initially designed for
the need for adjunctive stenting. Recent data have suggested the coronary circulation—was made available in larger diam-
significant reduction in restenosis with paclitaxel-coated bal- eters and was approved by the FDA for peripheral vascular
loons compared to POBA for infrapopliteal arteries (9,10). use. The device consists of four microsurgical blades bonded to
Peripheral drug-coated balloons have entered the market of the balloon. The cutting balloon has shown promise, in treat-
some European countries but are not approved by the U.S. ment of anastomotic or intragraft lesions within saphenous vein
Food and Drug Administration (FDA). bypass grafts. A recent study compared POBA with the cutting
balloon as the primary treatment for failing infrainguinal
Cryoplasty bypass grafts in 36 patients; the initial technical success was
Cryoplasty is a technique that combines balloon angioplasty higher with the cutting balloon (74% vs. 82%). In addition, the
and cold therapy. Cooling of the balloon is achieved by the use primary patency rate at 12 months was 36% for POBA and 50%
of liquid nitrous oxide as the balloon inflation media rather for cutting balloon angioplasty (12). Ansel et al. reported cut-
than the usual mixture of contrast and saline. The combination ting balloon angioplasty results in 73 patients with critical limb
is proposed to reduce the incidence of restenosis by inhibition ischemia. The procedure was successfully completed in all
of neointimal hyperplasia. Nonrandomized clinical studies patients, although 4% of cases required predilation with
have suggested improved patency associated with cryoplasty POBA. Severe intimal dissection or inadequate hemodynamic
compared to conventional angioplasty, but the benefit remains result necessitated adjunctive stenting in 20% of the proce-
to be adequately demonstrated (11). The PolarCath (Boston dures. There were no vessel perforations or surgical target
Scientific, Natick, Massachusetts, U.S.) is a commercially avail- vessel revascularization. After mean follow-up of one year,
able FDA-approved cryoplasty system for peripheral arterial 89.5% of threatened limbs were salvaged (13). The Scoring
use. The PolarCath system was evaluated in a prospective balloon (AngioSculpt, Inc., Fremont, California, U.S.) has been
multicenter registry of cryoplasty for the treatment of lesions developed for both coronary and peripheral use. They differ
in the SFA and popliteal artery. A total of 102 patients with from cutting balloons in that they have rectangular spiral struts.
stenoses or occlusions up to 10 cm in length were enrolled at 15 Scoring balloons come in sizes ranging from 2 to 6 mm in
centers in the United States. Procedural success was 94%, with diameter with 20 and 40 mm lengths.
[23/6/010/11:13:38] [610–624]
Nitinol Stents sites with high degrees of vessel torsion such as distal SFA, and
Nitinol is an alloy composed of nearly equal parts of nickel and popliteal artery to limit stent fracture. Table 49.5 provides
titanium. Nitinol has two unique characteristics: great elasticity information on a variety of self-expanding and balloon-expand-
and better conformability and thermal shape memory. These able stents for peripheral use. Drug-eluting stents (DES) do
properties make nitinol stents resistant to compression at body intuitively represent the logical answer to restenosis, although
temperature and help them to resist external deformation (4). data from The Sirolimus Coated Cordis SMART Nitinol Self-
The IntraCoil stent was one of the first nitinol stents approved Expandable Stent for the Treatment of SFA Disease (SIROCCO)
by the FDA. Ansel et al. studied 93 patients with obstructive trials that randomized patients with SFA disease to sirolimus-
femoropopliteal artery disease up to 15 cm in length treated coated SMART stents or noncoated SMART stents found no
with the IntraCoil stent. Nine-month follow-up revealed that statistical differences in terms of restenosis at 18 months (20,21).
78% of patients remained free of major adverse clinical events There are two ongoing DES trials using different drugs and
and that 82% had not required target lesion revascularization drug-release concepts. Currently, one study is investigating the
(14). The SMART stent (Cordis, Bridgewater, New Jersey, U.S.) paclitaxel-eluting Zilver PTX stent (Cook Medical, Inc., Bloo-
was evaluated in a prospective study involving 137 lower limbs mington, Indiana, U.S.). A second DES platform for SFA treat-
in 122 patients with chronic limb ischemia. The mean lesion ment is addressed in the STRIDEs registry using longer elusion
length was 12 cm and 57% of the lesions were at least 10 cm in times with everolimus from the DYNALINK-E stent (Abbott
length. Technical success was achieved in 98% of cases, and Vascular, Diegem, Belgium). Both studies and platforms were
there were no stent occlusions at 30 days. Primary patency designed to correct flaws in drug dose, and release kinetics
rates, determined by duplex ultrasound, were 76% and 60% at believed to be responsible for the disappointing results in the
one and two years, respectively (15). Schillinger and colleagues SCIROCCO trials. The application of DES for infrapopliteal
randomized 104 patients with severe claudication caused by lesions has recently shown more promise than for infrainguinal
stenosis or occlusion of the SFA to either primary stent implan- lesions. The combined results of four observational studies
tation with the Absolute stent (Abbott) (51 patients) or angio- evaluating the coronary Cypher stent (Cordis) for below-the-
plasty (53 patients). Bailout stenting was performed in 32% of knee procedures resulted in 100% freedom from major ampu-
patients in the angioplasty group mostly due to suboptimal tation, 96% limb salvage rates, and only a 4% need for target
result. At six months, the rate of restenosis on angiography was vessel revascularization, which compared well with the histor-
24% in the stent group and 43% in the angioplasty group ical surgical revision rates as high as 49% (22–25).
(P ¼ 0.05). Duplex ultrasonography at 12 months revealed Although balloon-expandable stents are commonly used
restenosis 50% in the stent group to be 37% compared to 63% for distal aortic stenosis and ostial iliac disease, self-expanding
in the angioplasty group (P ¼ 0.01). These outcomes translated nitinol stents have been shown to result in primary patency
into a significantly improved walking distance at both 6 and rates of greater than 90% at 12 months. The Cordis Randomized
12 months for the stent group (16). At 24 months, a persistent Iliac Stent Project-US (CRISP-US) trial evaluated the SMART
trend toward better treadmill walking capacity, higher ankle nitinol self-expanding stent and the stainless steel Wallstent
brachial index, and a reduction in reintervention for the patients (Boston Scientific) for treating iliac artery disease after subopti-
who underwent primary stenting was observed (17). mal POBA. Primary patency at 12 months was 95% with the
The Femoral Artery Stenting Trial (FAST) assessed the SMART stent and 91% for those lesions treated with the Wall-
Bard Luminex 3 Vascular Stent (C.R. Bard, Inc., Murray Hill, stent. Acute procedural success rate was higher in the SMART
New Jersey, U.S.) versus POBA. A total of 244 patients with a stent group (98.2% vs. 87.5%; P ¼ 002).
single-SFA lesion and chronic limb ischemia were randomized
to implantation of a single Luminex 3 stent (123 patients) or Nitinol Stent Grafts and Covered Stents
POBA (121 patients). At 12 months there was no significant The Viabahn endoprosthesis (W.L. Gore & Associates, Inc.,
difference in the primary endpoint of ultrasound-assessed Flagstaff, Arizona, U.S.) is a self-expanding helical nitinol stent
binary restenosis or target lesion revascularization rates (18). mounted to the outside surface of a tube of expanded poly-
The lack of benefit observed may be related to the relative short tetrafluoroethylene. The Viabahn endoprosthesis, delivered
lesion length in the trial (mean of 45 mm) compared to the data through a 7- or 8-Fr sheath, is currently approved by the FDA
reported by Schillinger et al. (16,17). In a study conducted by for use in patients with symptomatic superficial femoral arterial
Scheinert and colleagues, a total of 121 limbs in 93 patients were lesions with reference vessel diameters of 4.8 to 7.5 mm (4). The
treated by implantation of self-expanding nitinol stents. Viabahn stent has been tested in a randomized prospective
SMART stents (Cordis) were used in 52, SelfX stents (Abbott, study and was compared to surgical femoral-to-above knee
Beringen, Switzerland) in 24, and Luminex stents (C.R. Bard, popliteal artery bypass with synthetic graft material. A total of
Inc.) in 45 limbs, respectively (19). The mean length of the 100 limbs in 86 patients were treated. Follow-up evaluation with
stented segment was 15.7 cm. The patients were investigated by ankle-brachial indices and color flow duplex sonography imag-
X ray after a mean follow-up time of 11 months. Overall, stent ing were performed at 3, 6, 9, and 12 months after treatment.
fractures were detected in 45 of 121 treated limbs (37%). Clin- Although underpowered to be conclusive, no statistical differ-
ical outcome was significantly affected by the presence of stent ence was found in the primary patency, secondary patency, or
fracture with primary patency rates at 12 months of 41% in reintervention between the two treatment groups (26). Stent
limbs with a documented fracture versus 84% in fracture-free grafts and covered stents may be useful for aneurismal disease,
lesions (P < 0.0001) (19). given their ability to exclude the vessel wall from the lumen as
These data have to be taken into account when consid- has been reported for popliteal aneurysms (26,27). Additionally,
ering the use of nitinol stents. Despite the great initial angio- the Viabahn endoprosthesis is very effective for bailout SFA
graphic results associated with the use of these devices, the best perforations as seen in Figure 49.4. The iCAST (Atrium Medical
approach may be to limit the use of nitinol stents to lesions with Corp., Hudson, New Hampshire, U.S.) is balloon-expandable
suboptimal endovascular results as well as avoiding stenting of lower profile covered stent with diameters ranging from 5 to
[23/6/010/11:13:38] [610–624]
Figure 49.4 (A) SFA perforation (arrows) post balloon angioplasty with 4 100 mm2 balloon; (B) post 6 150 mm2 Viabahn stent fully
covering lesion and perforation. Abbreviation: SFA, superficial femoral artery.
10 mm, which can be delivered through a 6- and 7-Fr sheath. tibial vessels (Table 49.6). Limitations to the device have been its
The iCAST may be preferable for coverage of iliac aneurysms inability to cut calcified plaques, needing to use more than one
and in emergent cases of iliac perforations secondary to its device if treating varying size vessels in the same patient, and
crossing profile, precise placement that a balloon-expandable the risk of distal embolization requiring distal protection espe-
stent affords, and having the character of being able to expand cially for patients with limited patent runoff vessels.
to the desired diameter with post dilatation. The Diamondback 3608 Orbital Atherectomy System is a
promising new device for treating symptomatic PAD within
Debulking Devices the major and branch arteries of the leg. The device differs from
Atherectomy has been used to treat PAD in the past with mixed other atherectomy technologies by its unique orbital action to
results. More recently, there has been renewed interest in remove plaque and the ability to increase treatment diameter
debulking devices with the development of the SilverHawk by increasing orbital speed. Orbital Atherectomy System for
(Fox Hollow Technologies, Redwood City, California, U.S.), as Treating Peripheral vascular Stenosis (OASIS), a prospective
a new excisional atherectomy system, and the Diamondback multicenter clinical study, was conducted to evaluate the effi-
360 Orbital Atherectomy System (Cardiovascular System Inc., cacy and safety of this system; from September 1 through
St. Paul, Minnesota, U.S.), as a rotational atherectomy device. December 31, 2007, more than 350 cases with the Diamondback
Both devices are FDA approved. Although there are no random- 3608 were documented. Results revealed low rates of dissection
ized trials comparing excisional atherectomy to balloon angio- (2%), perforation (2.3%), and embolism (2%). Advantages for
plasty or stenting, there are several registries and single-center this device over the SilverHawk are its ability to cut through
experiences with SilverHawk system. Treating Peripherals with calcified lesions and the ability to increase cutting size by
SilverHawk: Outcomes Collection Registry (TALON) enrolled increasing rotational speed, thus potentially limiting the num-
601 consecutive patients in 19 institutions. A total of 1258 ber of catheters needed per case. It is hampered by its inability
symptomatic lower extremity atherosclerotic lesions were to cut soft plaque and tendency for distal embolization without
treated with mean lesion lengths of 63 mm above the knee and the potential for distal protection given its need for a dedicated
69 mm below the knee. The primary endpoints of the study were wire (Table 49.6).
target lesion revascularization (TLR) at 6 and 12 months. Proce-
dural success was 98% and the 6- and 12-month freedom of TLR Excimer Laser Atherectomy
rates were 90% and 80%, respectively. Predictors of TLR were a Excimer laser–assisted angioplasty (ELA) for the treatment of
history of MI or coronary revascularization, increasing Ruther- PAD has been commercially available in Europe since 1994. The
ford category, and lesion length. Multiple single center series technique is based on intense bursts of ultraviolet light in short
have also reported good initial success with similar mid- and pulse durations. The advantage of ELA lies in the ability to
long-term outcomes with this device within the femoropopliteal break molecular bonds directly by photochemical rather than
and infrapopliteal vessels (28–31). The advantage of the Silver- by pure heat, which limits continuous-wave hot-tip lasers. ELA
Hawk system is the ability to remove plaque with or without is most applicable in the treatment of long complex lesions—
the performance of adjunctive angioplasty. This treatment may calcified or noncalcified—and to facilitate crossing of CTO. The
be particularly useful for locations where stenting should be most advanced system used is the TURBO elite laser system
avoided such as the distal SFA, popliteal, and CFA. The system (Spectranetics, Colorado Springs, Colorado, U.S.). When com-
comes in various sizes allowing treatment from the CFA to the bined with the TURBO-Booster guiding catheter, the elite laser
[23/6/010/11:13:38] [610–624]
can be used to create larger lumens for infrainguinal arteries. Pioneer catheter (21 CTO) or fluoroscopic-guided true lumen
Scheinert and colleagues analyzed data from 318 consecutive reentry using the Outback catheter (30 CTO) was successful in
patients, who underwent ELA of 411 SFA with chronic occlu- achieving true lumen reentry in all cases. During this study, the
sions averaging 19.4 6.0 cm in length. Initial and secondary time to reentry was routinely less than 10 minutes (36). The
crossing success were 82% and 90.5%, respectively. The pri- Outback LTD device can be delivered through a 6-Fr sheath and
mary patency at one year was 33.6% with the one-year assisted has a smaller crossing profile than the Pioneer catheter that
primary and secondary patency rates being 65.1% and 75.9%, requires a 7-Fr sheath. The Outback’s lower profile is achieved
respectively (32). In the Laser Angioplasty for Critical Limb by using a system of angiographic views to align the catheter
Ischemia (LACI) trial, a prospective registry at 14 sites in the prior to needle reentry, while the Pioneer has a bulkier ultra-
United States and Germany, 145 patients with 155 critically sound tip to help locate the true lumen prior to needle reentry
ischemic limbs were enrolled. At six-month follow-up, limb (37). Both devices are FDA approved for peripheral intervention.
salvage was achieved in 92% of surviving patients (33).
Rheolytic Thrombectomy
Devices for Chronic Total Occlusions Although the FDA has approved many mechanical thrombec-
The Frontrunner XP CTO Catheter (Cordis) enables controlled tomy devices for use in thrombosed hemodialysis grafts, only
crossing of CTO. It uses blunt microdissection to create a the AngioJet LF140 (Possis Medical, Inc., Minneapolis, Minne-
channel through the occlusion to facilitate wire placement. It sota, U.S.) is currently approved for use in peripheral arterial
features a crossing profile of 0.039 in. with actuating jaws that occlusive disease (38). The Angiojet rheolytic thrombectomy
open to 2.3 mm. The shapeable distal tip and effective torque system has been shown to be effective in the treatment of
control enhance maneuverability while supported by a 4.5-Fr acutely occluded infra-aortic native arteries and bypass grafts,
Micro Guide catheter. Once the lesion is crossed, the guide with the majority of acute thrombotic material being removed.
catheter is advanced into the true lumen and the Frontrunner is The rheolytic catheter has been used successfully in conjunction
removed and replaced with a guidewire up to 0.035 in. Mossop with catheter-directed thrombolysis in one series of 86 patients
et al. prospectively evaluated the technical success and safety of with acute and subacute limb-threatening ischemia. After pri-
controlled blunt microdissection for the treatment of resistant mary rheolytic thrombectomy was performed, secondary cath-
peripheral CTO. They enrolled 36 patients with 44 symptomatic eter–directed thrombolysis was performed in 50 patients,
CTO (2 terminal aortic, 24 iliac, 16 femoral, and 2 popliteal), yielding a high acute success rate with a reported six-month
which had previously failed conventional percutaneous revas- patency rate of 79% (4,39). Additional evidence has been
cularization. Procedural success was achieved in 91% of the 44 reported in a retrospective analysis by Ansel et al. in which
CTO (34). This device is most helpful when the proximal cap of 99 consecutive patients underwent rheolytic therapy for throm-
a CTO is refractory to initial guide wire penetration. botic occlusions in 80 native arteries or 19 bypass grafts. Com-
The CROSSER (FlowCardia, Inc., Sunnyvale, California, plete thrombus removal was accomplished in 71% of patients
U.S.), a novel device for recanalization of CTO, is on a monorail and partial in 22%. Mortality and amputation rates at 30 days
catheter that delivers vibrational energy to facilitate the cross- were 7.1% and 4.0%, respectively (40). Although the Angiojet
ing of occluded arteries. Although the CROSSER system has system likely provides a more forceful and complete thrombus
been proven safe and feasible in the totally occluded coronary aspiration, other more simple devices are available; Pronto V3
circulation with procedure success rate reaching 73% (35), there extraction catheter (Vascular Solutions, Minneapolis, Minne-
is no clinical trial to assess the safety of this system in PAD. The sota, U.S.), Export XT catheter (Medtronic Vascular, Santa Rosa,
peripheral system is advanced over a 0.018 in. guidewire until California, U.S.), Fetch (Possis Medical, Inc.), and the Diver CE
it reaches the lumen cap when the wire is retracted within the (ev3, Plymouth, Minnesota, U.S.). Similarly, simple straight
Crosser catheter. Once the catheter crosses the length of the end-hole catheters can be used for thrombus aspiration.
lesion and reenters the true lumen, the wire is then reintro-
duced into the true lumen and the catheter removed making
Embolic Protection in Lower Extremity Revascularization
way for advancement of adjunctive equipment. This device
Since initial introduction for distal protection during carotid
relies on remaining within the true lumen and therefore may
stenting, embolic protection devices (EPD) are now routinely
not be effective in long total occlusions. Its nitch may be in
used in coronary vein graft intervention as well as carotid artery
initial crossing of heavily calcified initial caps. Both devices are
stenting. These devices are either balloon based, GuardWire Plus
FDA approved.
(Medtronic Vascular), or filter-based, AngioGuard (Cordis),
FilterWireEX (Boston Scientific), NeuroShield (MedNova,
True-Lumen Reentry Devices Galway, Ireland), and the Spider (ev3, Plymouth). Although
The OutBack LTD reentry catheter (Cordis) was first evaluated in there are no randomized trials currently accessing EPD for
a series of 36 patients with peripheral—mainly iliac and femo- lower extremity revascularization, there have been studies that
ral—CTO initially approached with percutaneous controlled suggest these devices may be efficacious (41–43). The use of these
blunt microdissection. Of all successful cases, 35% (14 of 40) devices is limited in the lower extremitiy by their micropores—
required true-lumen reentry, which was successfully achieved in which are designed to retain cholesterol particles—and their
all cases with this first generation device (34). Additional evi- susceptibility to becoming overwhelmed with fibrinous debris
dence has been reported for the use of both the Outback and (41,44). Although routine use of EPD for lower extremity inter-
Pioneer catheter (Medtronic, Inc., Minneapolis, Minnesota, U.S.), ventions cannot be routinely recommended, there are specific
by Jacobs et al. in which 87 CTO in 58 iliac and 29 superficial cases in which protection may be efficacious. Consideration
femoral arteries were treated. In 24 (26%), the true lumen could should be given for EDP for any lesion proximal to a single-vessel
not be reentered by using standard catheter and wire techniques. runoff, SilverHawk atherectomy in heavily calcified lesions, and
Intravascular ultrasound-guided true lumen reentry using the prior to rheolytic thrombectomy.
[23/6/010/11:13:38] [610–624]
access in an antegrade manner. Infrapopliteal repair is most

commonly performed either in a retrograde approach from the
contralateral CFA or in an antegrade approach via the ipsi-
lateral CFA. Rarely the PT or the AT artery will be accessed if
the lesion cannot be crossed via either of the previously
mentioned access locations. Once the lesion is crossed via the
access in the PT or AT, the wire can be snared and externalized
through the larger sheaths in the ipsilateral or contralateral
CFA. Because of the workable length of interventional equip-
ment, a brachial approach can only be utilized for the most
proximal segments of the SFA. On rare occasions, the operator
may choose an axillary access to utilize the most workable
length possible if there are no other alternatives.
Endovascular Repair
Iliac and CFA intervention Endovascular repair of the
common iliac can be one of the most straightforward lower
extremity interventions, especially when the lesion does not
Figure 49.5 Access sites and different approaches to lower involve the bifurcation of the distal aorta. In this case, a retro-
extremities intervention. grade approach from the ipsilateral CFA with a 23-cm 6- to 7-Fr
sheath will provide adequate length and size for most interven-
tional equipment, as seen in Figure 49.3. In the absence of a
total occlusion, any straight floppy tip 0.035 in. wire can be
Clinical Aspects and Technique used. In many occasions, a 0.018 in. wire will also provide
Access sufficient support. The size of the common iliac artery ranges
Aortoiliac and common femoral repair Access for endo- from 7 to 10 mm in diameter. If the deployment of a stent has
vascular interventions will depend on the territory to be inter- been deemed necessary and the ostium of the distal aorta is not
vened upon and the access availability of the patient, as involved, a self-expanding stent may be the preferred device.
outlined in Figure 49.5. Most aortoiliac interventions are best Accordingly, its flexibility allows for better vessel conformity
approached from the ipsilateral CFA in a retrograde approach, and apposition in tortuous and ectatic vessels while allowing
as seen in Figure 49.3. This allows for good visualization while for over sizing without the risk of perforation or vessel rupture.
ensuring the most straightforward intervention. If the ipsilat- Once the stent is in place, it can be post-dilated up to the size of
eral CFA is occluded, consideration can be given to access of the stent chosen. Balloon-expandable stents are commonly used
the contralateral CFA as long as the lesion to be treated is not in when radial strength and precise location is needed such as the
the ostium or proximal segment of the common iliac, allowing aortic bifurcation and when reforming the distal aorta bifurca-
enough distance from the tip of the sheath and the lesion to tion with a kissing stent technique. In distinction from self-
adequately deliver the interventional equipment. Brachial expanding stents, balloon-expandable stents may be post-
access in this situation is a good alternative. Lesions at the dilated to larger sizes.
bifurcation of the common iliacs will require accessing both When performing iliac interventions, the operator must
CFAs for the deployment of simultaneous balloons and or be prepared for the rare but potentially lethal complication of
stents at the distal aorta reforming the bifurcation. In the case perforation. Once recognized a balloon must be rapidly placed
of occlusion of one or both CFA, access from a brachial artery across the lesion to occluded flow. Access of the contralateral
and contralateral CFA or both brachial arteries will allow the CFA is obtained, and an occlusion balloon is placed within the
same ability for simultaneous placement of stents at the aor- aorta and inflated. Once occlusion of the distal aorta is estab-
toiliac bifurcation. One limitation to using the brachial artery lished, the balloon in the ruptured iliac can be deflated and an
for common iliac interventions is the need for larger sheath size appropriately sized covered stent is then deployed to seal the
(7 Fr) when stent size exceeds 8 to –10 mm in diameter. Access rupture. A list of emergency equipment is found in Table 49.7.
for endovascular repair of the CFA most commonly will be
from the contralateral CFA in a retrograde approach or brachial
Table 49.7 Emergency Equipment for the Treatment of Iliac
if the contralateral CFA cannot be utilized.
Perforations
Femoral-popliteal and infrapopliteal repair A retrograde
crossover approach from the contralateral CFA access is com- Diameter Minimal
monly used to treat SFA lesions in any location along its course. (mm) sheath (Fr) Wire RX/OTW
This access is not recommended or not possible in the presence Occlusion balloon
of severe disease of the contralateral CFA, severe calcification Equilizer (Boston 27 14 0.035 OTW
or narrow angle of the aortoiliac bifurcation, and in the pres- Scientific)
ence of aortobifemoral or bi-iliac graft. In these cases, an 33 16 0.035 OTW
antegrade access from the ipsilateral CFA is an alternative as Covered stent
long as the proximal segment of the CFA is patent allowing Atrium (iCAST) 5 6 0.035 OTW
good seating of your sheath. When the proximal SFA is 6–8 7 0.035 OTW
diseased or occluded, the possible access options are either Viabahn (Gore) 6 6 0.035 OTW
through a retrograde ipsilateral popliteal approach or repairing 7–8 8 0.035 OTW
the CFA from a brachial access with subsequent ipsilateral CFA Abbreviations: RX, rapid exchange; OTW, over the wire.
[23/6/010/11:13:38] [610–624]
Figure 49.6 (A) Focal lesion in the distal common femoral artery and extending into the SFA and profunda femoris bifurcation (arrow);
(B) kissing balloon angioplasty into the SFA and profunda femoris artery through a 6-Fr 45-cm sheath using the retrograde crossover
technique for access; (C) final result after kissing balloon angioplasty. Abbreviation: SFA, superficial femoral artery.
CFA intervention must be accomplished with preservation of the distal SFA via collateral flow from the profunda femoris. In
future access and without affecting the profunda femoris. the absence of a CTO, the lesion in the SFA can usually be
Bearing these factors in mind stents should be avoided and wired with any 0.035 in. wire. Predilation will allow trouble-
either stand-alone angioplasty, as seen in Figure 49.6, or athe- free crossing with the appropriately sized self-expanding stent.
rectomy, as seen in Figure 49.7, should be the modalities of Self-expanding stents are usually sized one vessel size up from
choice. Either a retrograde approach from the contralateral CFA the estimated vessel size. Post-dilation of the stent is then
or a brachial artery access is most commonly used. performed with a one-to-one balloon to vessel size ratio. Care
SFA intervention SFA repair is most commonly per- must be given in heavily calcified vessels not to overdilate,
formed from the contralateral CFA with the use of a 45-cm 6- to which may result in vessel rupture. Patient discomfort during
7-Fr sheath. The sheath is advanced into the CFA or into the balloon inflation is commonly used to detect when the maximal
proximal SFA of the affected side, if the vessel will allow such size has been reached. In locations within the SFA where stent
placement. The operator may desire placement of the tip of the placement is not recommended (distal to the Adductor Hiatus),
sheath prior to the takeoff of the profunda femoris in the setting SilverHawk atherectomy is a good option. In heavily
of a CTO of the SFA. This will allow for better visualization of calcified lesions, Diamondback atherectomy or ELA should be
[23/6/010/11:13:38] [610–624]
Figure 49.7 (A) Noncalcified high-grade CFA stenosis (arrow); (B) post atherectomy using SilverHawk MS system. Abbreviation: CFA,
common femoral artery.
Figure 49.8 Subintimal dissection of CTO. Abbreviation: CTO, chronic total occluded vessel.
considered. Both atherectomy and ELA outcomes may be plane. Maintaining the smallest dissection plane possible will
enhanced with post-atherectomy low atmosphere balloon increase the likelihood of reentering the true lumen with your
angioplasty. Cryoplasty, cutting, and scoring balloons also guidewire. Once the distal cap is reached, the wire is then
remain options in this location. In the presence of a CTO, retracted into the catheter and redirected without the loop in
there are various techniques and devices at the operator’s attempts to cross back into the true lumen, as seen in Fig-
disposal. For long occlusions subintimal angioplasty has ure 49.8. Typically, reentry into the true lumen is the most time
become the standard approach to revascularization. A stiff consuming and tedious aspect of treating a CTO. Reentry
hydrophilic wire is advanced the proximal cap with the aid catheters such as the Outback or Pioneer have shown remark-
of a backup catheter. The wire is then used to probe the cap able success if reentry into the true lumen is not possible
until a small loop is formed in the subintimal space. The with standard catheter and guidewire technique. When used
backup catheter is then advanced forward, keeping the wire to cross CTO, the Turbo elite laser catheter can be advanced
loop as small as possible limiting the size of the dissection in proximity to the initial occlusive cap and activated for 5 to
[23/6/010/11:13:38] [610–624]
Table 49.8 Devices for Use in CTO Endovascular Intervention should be placed as close to the lesion as possible to ensure
adequate visualization either from a retrograde or anterograde
Minimal CFA approach. The simplest approach is POBA with prolonged
sheath (Fr) Wire RX/OTW
inflation times. Cryoplasty and cutting balloon devices may
CTO crossing devices have theoretical advantages over POBA, but data has not been
Frontrunner (Cordis) 6 NA NA convincing (47,48). Long diffuse or bifurcating lesions may be
Crosser 14S (Flow Cardia) 5 0.014 RX better suited for atherectomy devices (SilverHawk and Dia-
Crosser 18 (Flow Cardia) 6 0.018 RX
mondback) with or without adjunctive balloon angioplasty, as
Reentry devices seen in Figure 49.10. An area currently being investigated is the
Outback LTD (Cordis) 6 0.014a RX use of coronary DES that have been shown to markedly reduce
Pioneer (Medtronic) 7 0.014a RX
restenosis compared to bare metal stents for bailout purposes in
a
Ironman or Grandslam wire. infrapopliteal vessels (49). Should these results be reproduced
and should longer and more deliverable drug-eluting devices
come to the market, stenting may become the therapeutic
10 seconds in a trial to penetrate the fibrous cap. The guidewire modality of choice for below-the-knee interventions, particu-
is then used to probe antegrade to find a channel through the larly for focal stenosis or occlusions. It is important to under-
occlusion. This process is repeated until the entire length of the score that even a temporary patency—for example, following
CTO is crossed (4). The Frontrunner and the Crosser 18 be used angioplasty—may be sufficient to allow for wound healing in
in situations when the proximal cap cannot be penetrated as patients with foot ulcers. On a broader perspective, it seems
described earlier. Either device can then be continued if it likely that endovascular repair for infrapopliteal lesions will
passes easily through the occlusion. Once across the distal soon become standard of care (50,51).
cap, full-length angioplasty is performed followed typically
by self-expanding stent implantation. CTO equipment and
compatibility is listed in Table 49.8. UPPER EXTREMITY REVASCULARIZATION
Popliteal intervention Similar to CFA interventions In contrast to lower extremity disease, upper extremity vascular
operators should avoid stenting the popliteal artery due to disease is uncommon and usually manifests with arm claudi-
the increase likelihood of stent fracture. The Supera stent cation, rest pain, and ischemic ulcerations. Rare manifestations
(IDev Technologies, Houston, Texas, U.S.) may expand the of a subclavian stenosis are vertebrobasilar insufficiency if the
role of bailout stenting in this location due to its superior contralateral vertebral artery is diseased/occluded and angina
flexibility and resistance to fracture. The Viahban stent graft in a patient with previous mammary artery used as a conduit
has been used for coverage of popliteal aneurysms exceeding for CABG. The most common cause of large artery disease of
2.0 cm in diameter. A critical limitation of the Viahban in this the upper extremity is atherosclerosis. The involvement is
location has been the occurrence of stent thrombosis following typically limited to the proximal segments of the innominate
prolonged knee flexure. Patients receiving a Viabahn stent in and subclavian arteries, while more distal lesions are rare. The
the popliteal artery should be warned to avoid prolonged majority of the remaining cases of occlusive disease are asso-
excessive knee flexion. Favored modalities for treatment of ciated with vasculitis, post-radiation, and external compres-
atherosclerotic disease of the popliteal artery remain Silver- sion. Goals of revascularization of the upper extremity include
Hawk atherectomy, balloon angioplasty, cryoplasty, cutting relief from claudication, prevention of digit or limb loss, pro-
balloon angioplasty, ELA, and Diamondback orbital atherec- tection of antegrade flow to the internal mammary artery after
tomy for calcified lesions. Figure 49.9 illustrates Diamondback CABG and of arterial-venous shunts for hemodialysis. Histor-
atherectomy followed by low atmosphere balloon angioplasty ically, upper extremity ischemic disease involving the large
of the popliteal artery. vessels was managed surgically using bypass or endarterec-
Infrapopliteal intervention Historically, revasculariza- tomy techniques. Currently, endovascular repair is the most
tion of the infrapopliteal vasculature has been open surgical accepted modality of treatment with data to support the lack of
and reserved to critical limb ischemia and limb salvage. One difference in long-term patency rates compared to surgery (52).
reason for this conservative strategy has been the poor long- More recently, Bates et al. reported a success rate of 97% and
term patency of surgical grafts and also of the initial endovas- one-year patency of 96% in 89 patients who underwent sub-
cular experience, the latter characterized by a high restenosis clavian artery stenting (53).
rate (45). Another major factor limiting an aggressive endovas-
cular approach has been the fear to compromise a situation
already critical with the risk of a subsequent amputation. Anatomic Considerations
However with advances in technology endovascular repair The proximal segments of the large arteries are seen in Fig-
appears to have many advantages over open surgical repair ure 49.11. The primary anatomical consideration when stenting
(2). The results of the Bypass versus Angioplasty in Severe the proximal left subclavian artery is the location of the left
Ischemia of the Leg (BASIL) trial revealed that endovascular vertebral artery. Appropriate angulations should be performed
repair compared to open repair for critical limb ischemia was to see the true ostium of the vertebral artery prior to stent
associated with lower morbidity, length of hospital stay, and implantation. The majority of cases will allow a landing zone
necessity for high intensity units—all resulting in lower costs. prior to the vertebral origin. When there is no landing zone
Overall clinical outcomes in the Basil trial were similar between prior to the vertebral artery, the operator must be aware of
the two groups with respect to amputation-free survival (46). the consequences of jailing or shifting plaque into the vertebral
Most below-the-knee vessels are 2 to 5 mm in diameter and artery. In this case, to avoid compromise of a dominant or sole
therefore best suited for 0.014 in. platform devices that are vertebral artery, a (subselective) angiography of both vertebral
commonly employed for coronary intervention. The sheath arteries—including intracranial views—is recommended.
[23/6/010/11:13:38] [610–624]
Figure 49.9 (A) High-grade stenosis of the popliteal artery; (B) post atherectomy with 2.25 Diamondback catheters in the popliteal artery
(arrow); (C) low-pressure balloon inflation after Diamondback atherectomy the popliteal artery (arrow); (D) final result after Diamondback
atherectomy.
When approaching the right system, the operator must ade- right internal carotid artery during repair of the right innomi-
quately visualize the bifurcation of the right common carotid nate may be considered for bulky atherosclerotic lesions.
and right vertebral arteries when repairing the innominate and
right subclavian artery, respectively. Typically, the bifurcation
of the right common carotid and innominate artery is best seen Access and Technique
in the RAO projection. In the setting of significant atheroscler- Access for endovascular repair of the innominate or suclavian
otic burden, distal protection within the vertebral system has artery can be via the common femoral, brachial, or radial
been advocated by some, although the risk of stroke is small arteries. The CFA approach will permit the use of either a
potentially due to delayed reversal of vertebral artery blood sheath or guide catheter. When using a sheath the operator can
flow from retrograde to antegrade following angioplasty of the either wire the lesion through a diagnostic catheter and replace
subclavian artery (54,55). Similarly, distal protection within the the entire system with a sheath (usually 6–7 Fr and 90 cm in
[23/6/010/11:13:38] [610–624]
Figure 49.10 (A) Bifurcation of the popliteal artery into the anterior tibial artery and tibioperoneal (TP) trunk revealing a totally occluded TP
trunk (arrowhead) and severe disease in the AT (arrow); (B) 1.25 Diamondback catheter advanced for multiple runs in the AT; (C) final result
in the PT trunk and AT after atherectomy and low-atmosphere balloon inflation. Abbreviations: AT, anterior tibial; PT, posterior tibial.
length) or more simply telescope the diagnostic catheter strength and precise placement. If the lesion is located aorto-
through the sheath and once wired advance the sheath into ostially, the stent should protrude 1 to 2 mm into the aorta
the appropriate location over the diagnostic catheter. Guiding to have a complete coverage of the lesion. Although most
catheters are limited by their smaller internal diameter when investigators believe that no distal protection of the vertebral
compared to that of similar sized sheath. Sheaths are routinely artery is required at the time of proximal upper extremity
used during brachial (6–7 Fr, 45 cm) and radial (6 Fr, 65–90 cm) intervention, thoughtful consideration must be given to pro-
access. Once the lesion has successfully been crossed with a tecting the right internal carotid for bulky innominate lesions.
standard 0.035 in. wire, balloon predilatation can be used to Protection of the right internal carotid can be accomplished via
gage size and length prior to stent implantation. Subclavian access and deployment of the protection device through an
arteries usually accommodate 7 to 10 mm diameter stents, additional sheath in the right brachial or radial artery while
while innominate arteries may require larger devices. primarily working through a sheath from the CFA to repair the
Balloon-expandable stents are primarily used for their radial innominate.
[23/6/010/11:13:38] [610–624]
8. Jeans WD, Armstrong S, Cole SE, et al. Fate of patients undergoing

transluminal angioplasty for lower-limb ischemia. Radiology
1990; 177:559–564.
9. Tepe G, Zeller T, Albrecht T, et al. Local delivery of paclitaxel to
inhibit restenosis during angioplasty of the leg. N Engl J Med
2008; 358:689–699.
10. Werk M, Langner S, Reinkensmeier B, et al. Inhibition of reste-
nosis in femoropopliteal arteries: paclitaxel-coated versus
uncoated balloon: femoral paclitaxel randomized pilot trial. Cir-
culation 2008; 118:1358–1365.
11. Laird J, Jaff MR, Biamino G, et al. Cryoplasty for the treatment of
femoropopliteal arterial disease: results of a prospective, multi-
center registry. J Vasc Interv Radiol 2005; 16:1067–1073.
12. Vikram R, Ross RA, Bhat R, et al. Cutting balloon angioplasty
versus standard balloon angioplasty for failing infra-inguinal vein
grafts: comparative study of short- and mid-term primary patency
rates. Cardiovasc Intervent Radiol 2007; 30:607–610.
Figure 49.11 Anatomy of the aortic arch and the great vessels of 13. Ansel GM, Sample NS, Botti IC Jr., et al. Cutting balloon angio-
the neck. plasty of the popliteal and infrapopliteal vessels for symptomatic
limb ischemia. Catheter Cardiovasc Interv 2004; 61:1–4.
14. Ansel GM, Botti CF Jr., George BS, et al. Clinical results for the
training-phase roll-in patients in the Intracoil femoralpopliteal
stent trial. Catheter Cardiovasc Interv 2002; 56:443–449.
CONCLUSIONS AND FUTURE DEVELOPMENTS 15. Mewissen MW. Self-expanding nitinol stents in the femoropopli-
Endovascular repair can be accomplished today with success in teal segment: technique and mid-term results. Tech Vasc Interv
the majority of cases throughout the upper and lower extremity Radiol 2004; 7:2–5.
arterial bed with a low complication rate. Advances in technol- 16. Schillinger M, Sabeti S, Loewe C, et al. Balloon angioplasty versus
ogy and skills will further expand the proportion of patients implantation of nitinol stents in the superficial femoral artery.
undergoing endovascular interventions compared to open sur- N Engl J Med 2006; 354:1879–1888.
17. Schillinger M, Sabeti S, Dick P, et al. Sustained benefit at 2 years of
gical procedures. Current major limitations of this technique,
primary femoropopliteal stenting compared with balloon angio-
such as fracture and restenosis of self-expanding stents in the plasty with optional stenting. Circulation 2007; 115:2745–2749.
femoropopliteal circulation and the high reocclusion rate in 18. Krankenberg H, Schluter M, Steinkamp HJ, et al. Nitinol stent
below-the-knee procedures need to be resolved. In this respect, implantation versus percutaneous transluminal angioplasty in
the role of DES and balloons is currently being investigated, but superficial femoral artery lesions up to 10 cm in length: the
the pathway appears to be more complex than in the coronary femoral artery stenting trial (FAST). Circulation 2007; 116:
tree. Programs addressing best medical treatment and early 285–292.
disease detection should be implemented to prevent the devel- 19. Scheinert D, Scheinert S, Sax J, et al. Prevalence and clinical impact
opment of vasculopathy that results in limb loss. of stent fractures after femoropopliteal stenting. J Am Coll Cardiol
2005; 45:312–315.
20. Duda SH, Bosiers M, Lammer J, et al. Sirolimus-eluting versus
REFERENCES bare nitinol stent for obstructive superficial femoral artery disease:
1. Selvin E, Erlinger TP. Prevalence of and risk factors for peripheral the SIROCCO II trial. J Vasc Interv Radiol 2005; 16:331–338.
arterial disease in the United States: results from the National 21. Duda SH, Pusich B, Richter G, et al. Sirolimus-eluting stents for
Health and Nutrition Examination Survey, 1999–2000. Circulation the treatment of obstructive superficial femoral artery disease: six-
2004; 110:738–743. month results. Circulation 2002; 106:1505–1509.
2. Hirsch AT, Haskal ZJ, Hertzer NR, et al. ACC/AHA Guidelines 22. Bosiers M, Deloose K, Verbist J, et al. Percutaneous transluminal
for the Management of Patients with Peripheral Arterial Disease angioplasty for treatment of ’’below-the-knee’’ critical limb ische-
(lower extremity, renal, mesenteric, and abdominal aortic): a mia: early outcomes following the use of sirolimus-eluting stents.
collaborative report from the American Associations for Vascular J Cardiovasc Surg (Torino) 2006; 47:171–176.
Surgery/Society for Vascular Surgery, Society for Cardiovascular 23. Commeau P, Barragan P, Roquebert PO. Sirolimus for below the
Angiography and Interventions, Society for Vascular Medicine knee lesions: mid-term results of SiroBTK study. Catheter Cardi-
and Biology, Society of Interventional Radiology, and the ovasc Interv 2006; 68:793–798.
ACC/AHA Task Force on Practice Guidelines (writing committee 24. Scheinert D, Ulrich M, Scheinert S, et al. Comparison of sirolimus-
to develop guidelines for the management of patients with periph- eluting vs. bare-metal stents for the treatment of infrapopliteal
eral arterial disease)—summary of recommendations. J Vasc obstructions. EuroIntervention 2006; 2:169–174.
Interv Radiol 2006; 17:1383–1397; quiz 98. 25. Siablis D, Karnabatidis D, Katsanos K, et al. Sirolimus-eluting
3. Norgren L, Hiatt WR, Dormandy JA, et al. Inter-society consensus versus bare stents after suboptimal infrapopliteal angioplasty for
for the management of peripheral arterial disease (TASC II). J Vasc critical limb ischemia: enduring 1-year angiographic and clinical
Surg 2007; 45(suppl S):S5–S67. benefit. J Endovasc Ther 2007; 14:241–250.
4. Rogers JH, Laird JR. Overview of new technologies for lower 26. Burger T, Meyer F, Tautenhahn J, et al. Initial experiences with
extremity revascularization. Circulation 2007; 116:2072–2085. percutaneous endovascular repair of popliteal artery lesions using
5. Capek P, McLean GK, Berkowitz HD. Femoropopliteal angioplasty. a new PTFE stent-graft. J Endovasc Surg 1998; 5:365–372.
Factors influencing long-term success. Circulation 1991; 83:I70–180. 27. Tielliu IF, Verhoeven EL, Prins TR, et al. Treatment of popliteal
6. Dorrucci V. Treatment of superficial femoral artery occlusive artery aneurysms with the Hemobahn stent-graft. J Endovasc Ther
disease. J Cardiovasc Surg (Torino) 2004; 45:193–201. 2003; 10:111–116.
7. Henriksen LO, Jorgensen B, Holstein PE, et al. Percutaneous 28. Kandzari DE, Kiesz RS, Allie D, et al. Procedural and clinical
transluminal angioplasty of infrarenal arteries in intermittent outcomes with catheter-based plaque excision in critical limb
claudication. Acta Chir Scand 1988; 154:573–576. ischemia. J Endovasc Ther 2006; 13:12–22.
[23/6/010/11:13:38] [610–624]
29. Matsi PJ, Manninen HI, Vanninen RL, et al. Femoropopliteal 43. Wholey MH, Toursarkissian B, Postoak D, et al. Early experience
angioplasty in patients with claudication: primary and secondary in the application of distal protection devices in treatment of
patency in 140 limbs with 1-3-year follow-up. Radiology 1994; peripheral vascular disease of the lower extremities. Catheter
191:727–733. Cardiovasc Interv 2005; 64:227–235.
30. Zeller T, Rastan A, Schwarzwalder U, et al. Midterm results after 44. Konig CW, Pusich B, Tepe G, et al. Frequent embolization in
atherectomy-assisted angioplasty of below-knee arteries with use peripheral angioplasty: detection with an embolism protection
of the Silverhawk device. J Vasc Interv Radiol 2004; 15:1391–1397. device (AngioGuard) and electron microscopy. Cardiovasc Inter-
31. Zeller T, Rastan A, Sixt S, et al. Long-term results after directional vent Radiol 2003; 26:334–339.
atherectomy of femoro-popliteal lesions. J Am Coll Cardiol 2006; 45. Soder HK, Manninen HI, Jaakkola P, et al. Prospective trial of
48:1573–1578. infrapopliteal artery balloon angioplasty for critical limb ischemia:
32. Scheinert D, Laird JR Jr., Schroder M, et al. Excimer laser-assisted angiographic and clinical results. J Vasc Interv Radiol 2000;
recanalization of long, chronic superficial femoral artery occlu- 11:1021–1031.
sions. J Endovasc Ther 2001; 8:156–166. 46. Adam DJ, Beard JD, Cleveland T, Basil Trial Participants. Bypass
33. Laird JR, Zeller T, Gray BH, et al. Limb salvage following laser- versus angioplasty in severe ischaemia of the leg (BASIL): Multi-
assisted angioplasty for critical limb ischemia: results of the LACI centre, randomised, controlled trial. J Vasc Surg 2006; 44:430.
multicenter trial. J Endovasc Ther 2006; 13:1–11. 47. Arain SA, White CJ. Endovascular therapy for critical limb ische-
34. Mossop P, Cincotta M, Whitbourn R. First case reports of con- mia. Vasc Med 2008; 13:267–279.
trolled blunt microdissection for percutaneous transluminal 48. Das T, McNamara T, Gray B, et al. Cryoplasty therapy for limb
angioplasty of chronic total occlusions in peripheral arteries. salvage in patients with critical limb ischemia. J Endovasc Ther
Catheter Cardiovasc Interv 2003; 59:255–258. 2007; 14:753–762.
35. Melzi G, Cosgrave J, Biondi-Zoccai GL, et al. A novel approach to 49. Siablis D, Kraniotis P, Karnabatidis D, et al. Sirolimus-eluting
chronic total occlusions: the crosser system. Catheter Cardiovasc versus bare stents for bailout after suboptimal infrapopliteal
Interv 2006; 68:29–35. angioplasty for critical limb ischemia: 6-month angiographic
36. Jacobs DL, Motaganahalli RL, Cox DE, et al. True lumen re-entry results from a nonrandomized prospective single-center study.
devices facilitate subintimal angioplasty and stenting of total J Endovasc Ther 2005; 12:685–695.
chronic occlusions: Initial report. J Vasc Surg 2006; 43:1291–1296. 50. Bosiers M, Hart JP, Deloose K, et al. Endovascular therapy as the
37. Saket RR, Razavi MK, Padidar A, et al. Novel intravascular ultra- primary approach for limb salvage in patients with critical limb
sound-guided method to create transintimal arterial communica- ischemia: experience with 443 infrapopliteal procedures. Vascular
tions: initial experience in peripheral occlusive disease and aortic 2006; 14:63–69.
dissection. J Endovasc Ther 2004; 11:274–280. 51. Kudo T, Chandra FA, Kwun WH, et al. Changing pattern of
38. Kasirajan K, Haskal ZJ, Ouriel K. The use of mechanical throm- surgical revascularization for critical limb ischemia over
bectomy devices in the management of acute peripheral arterial 12 years: endovascular vs. open bypass surgery. J Vasc Surg
occlusive disease. J Vasc Interv Radiol 2001; 12:405–411. 2006; 44:304–313.
39. Kasirajan K, Gray B, Beavers FP, et al. Rheolytic thrombectomy in 52. Rodriguez-Lopez JA, Werner A, Martinez R, et al. Stenting for
the management of acute and subacute limb-threatening ischemia. atherosclerotic occlusive disease of the subclavian artery. Ann
J Vasc Interv Radiol 2001; 12:413–421. Vasc Surg 1999; 13:254–260.
40. Ansel GM, George BS, Botti CF, et al. Rheolytic thrombectomy in 53. Bates MC, Broce M, Lavigne PS, et al. Subclavian artery stenting:
the management of limb ischemia: 30-day results from a multi- factors influencing long-term outcome. Catheter Cardiovasc Interv
center registry. J Endovasc Ther 2002; 9:395–402. 2004; 61:5–11.
41. Siablis D, Karnabatidis D, Katsanos K, et al. Outflow protection 54. Nasim A, Sayers RD, Bell PR, et al. Protection against vertebral
filters during percutaneous recanalization of lower extremities’ artery embolisation during proximal subclavian artery angio-
arterial occlusions: a pilot study. Eur J Radiol 2005; 55:243–249. plasty. Eur J Vasc Surg 1994; 8:362–363.
42. Suri R, Wholey MH, Postoak D, et al. Distal embolic protection 55. Ringelstein EB, Zeumer H. Delayed reversal of vertebral artery
during femoropopliteal atherectomy. Catheter Cardiovasc Interv blood flow following percutaneous transluminal angioplasty for
2006; 67:417–422. subclavian steal syndrome. Neuroradiology 1984; 26:189–198.
[2/7/010/20:55:16] [625–639]
50
Renal and mesenteric artery interventions

John H. Rundback and Robert Lookstein
RENAL ARTERY INTERVENTIONS an initial noninterventional strategy utilizing careful surveil-

Introduction lance for progressive renal disease may be justified.
Renal artery stenosis (RAS) is a common clinical condition. PTRA or PTRS is supported in patients with a hemody-
Atherosclerotic RAS (ARAS) is found in up to 75% of individ- namically significant RAS and an appropriate clinical condition
uals at autopsy (1) and is frequently coexistent with athero- for treatment (Table 50.1) (11,12). Intervention for accelerated,
sclerosis occurring in other vascular beds (2,3). Fibromuscular resistant, or malignant hypertension as well as hypertension
dysplasia (FMD) is a unique arteriopathy, often affecting the with an unexplained unilateral small kidney or with intoler-
renal arteries, and typically occurring in Caucasian females in ance to medication is considered by the ACC/AHA guidelines
their third to fifth decade of life, with a high predilection for to be a Class IIa (Level of Evidence: B) treatment. Similarly,
bilateral disease (4). Regardless of etiology, RAS may be an PTRS for RAS and progressive chronic kidney disease with
incidental finding, may be associated with or causative of bilateral RAS or RAS to a solitary functioning kidney is a
hypertension, or, in the case of ARAS, may contribute to Class IIa (Level of Evidence: B) treatment. Revascularization
declining renal function or cardiac syndromes including con- for patients with chronic renal insufficiency with unilateral
gestive heart failure (CHF) and angina pectoris (5). In addition, RAS is a Class IIb (Level of Evidence: C) therapy. In contrast,
there is some evidence that ARAS confers an increased risk of stenting for significant RAS and recurrent, unexplained CHF or
stroke independent of blood pressure control, myocardial sudden, unexplained pulmonary edema is a Class I (Level of
infarction (MI), and cardiovascular death (6), although no Evidence: C) intervention (12). Obviously, such guidelines will
interventional strategy has yet been proven beneficial in miti- undergo periodic reevaluation as increasing data are accumu-
gating this risk. The currently enrolling U.S. National Institute lated regarding the role and relative outcomes of PTRS and
of Health (NIH)-sponsored CORAL trial (Cardiovascular Out- other potential treatments (7,14).
comes in Renal Artery Lesions) will hopefully provide insights Prognostic factors for both a favorable and unfavorable
into the treatment of this condition (7). response to renal intervention have been described. Best res-
The endovascular management of RAS with percutane- ponders for hypertension include hyperreninemia (either basal
ous transluminal renal angioplasty (PTRA) and stenting (PTRS) or angiotensin-converting enzyme stimulated), shorter hyper-
was first described more than 20 years ago (8). While PTRA tension duration, higher baseline diastolic blood pressure, more
alone is generally effective for the treatment of FMD, PTRS has severe angiographic stenosis, measurable transstenotic gradient
emerged in the modern era as the standard of care for patients of at least 20 mmHg (15), and presence of FMD (16,17). Best
with ARAS. As such, suboptimal prior clinical trial data com- responders for renal function preservation are patients with
paring medical therapy to PTRA alone for the treatment of more severe stenosis and rapid declines in function prior to
ARAS are not informative regarding clinical decision making at intervention (18). In general, worse outcomes are seen in
the current time (9). Despite extensive observational evidence patients with marked target kidney atrophy (i.e., <7 cm pole-
of clinical benefits derived from PTRS in ARAS patients, ade- to-pole length), advanced and long-standing renal failure,
quately powered randomized studies comparing this technique longer duration of hypertension, proteinuria, and the presence
to observation alone, contemporary medical management, and of bilateral disease (19). Microcirculatory evidence, as mea-
surgical bypass are lacking (10). These issues will clearly be the sured by a duplex renal resistive index of >0.80, was found
focus of further study in the years ahead. in one study to be a negative outcome predictor for both
hypertension and renal function (20), although a larger study
did not confirm this finding (21). Finally, diabetes does not
Indications for Intervention necessarily portend a lack of blood pressure or creatinine
In most cases, intervention for an incidentally discovered RAS benefit, even in the presence of coexistent nephrosclerosis (22).
is not warranted, and established guidelines support the prac- Several new prognostic tools have been recently pro-
tice of screening for RAS only in individuals with clinical posed. An elevated serum biomarker level brain natriuretic
conditions supporting revascularization (Table 50.1) (11). Per- peptide level > 80 pg/mL has been identified in one study of
haps a single exception to this rule would be the identification 27 patients with multidrug-resistant hypertension and RAS as
of a severe (i.e., >80%) stenosis in an otherwise asymptomatic a predictor of good blood pressure response following
patient with a solitary functional kidney, due to an established revascularization (23). Renal fractional flow resistance (FFR)
risk of spontaneous arterial occlusion precluding subsequent has also been correlated with reduced blood pressure after
endovascular management (12,13). In all cases, individual case PTRS (24). Interestingly, it also appears that the FFR—defined
assessment regarding the risks and merits of the procedure as the intraprocedurally measured ratio of the mean pressure in
should be based on clinical, anatomic, hemodynamic, and the renal artery after maximum hyperemia (by intrarenally
comorbid factors. In patients with excessive procedural risk, administered papaverine vasodilation) to the aortic mean
[2/7/010/20:55:16] [625–639]
Table 50.1 Indications for Renal Intervention typically involves the mid to distal portions of the renal artery,
angioplasty alone is generally sufficient, with stenting reserved
I. Hypertension (86,88–90) for dissections (de novo or post-PTRA) or suboptimal
l Accelerated hypertension—sudden worsening of previously PTRA results (Fig. 50.1) (25). Balloon size is selected based
controlled hypertension
l
on the normal diameter of the renal artery proximal to the
Refractory hypertension—hypertension resistant to treatment
with at least 3 medications of different classes including FMD, rather than in areas of poststenotic dilation. Cases
a diuretic with complex associated aneurysm may require more complex
l Malignant hypertension—hypertension with coexistent management, including the use of stent grafts or selective
evidence of end organ damage, including left ventricular embolization.
hypertrophy, congestive heart failure, visual or neurological ARAS characteristically involves the ostium of the renal
disturbance, and/or advanced (grade IV) retinopathy artery, and PTRS has been clearly proven to be more efficacious
l Hypertension with a unilateral small kidney than PTRA alone (26). Consequently, most operators have
l Hypertension with intolerance to medication adopted a practice of primary or direct stenting. Despite the
II. Renal salvage fact that ostial calcification is often present, lesions are efface-
l Sudden unexplained worsening of renal function (91)
l
able with direct stent implantation in almost all cases. Stents
Impairment of renal function secondary to antihypertensive
treatment, particularly with an angiotensin-converting enzyme should be implanted so as to completely cover the culprit
inhibitor or angiotensin II receptor blocker (92,95) lesion, with 1 to 3 mm extension into the aorta in most cases.
l Renal dysfunction not attributable to another cause In scenarios in which remodeling of the aortorenal ostium
III. Cardiac disturbance syndromes proximal to the ARAS results in different medial positions of
l Recurrent “flash” pulmonary edema out of proportion to left the upper and lower edges of the renal artery origin, complete
ventricular impairment (2,93,94) stent coverage across the most medial edge is necessary.
l Unstable angina in the setting of significant RAS Positioning of renal artery stents requires that the treated
IV. Medication flexibility renal artery origin is seen in complete profile. Computed
l Clinical need to use ACEI or ARB (i.e., for treatment of tomography (CT) studies demonstrate that the right renal
proteinuria or cardiac dysfunction) that would be otherwise
artery is often best imaged from a straight anterior-posterior
contraindicated due to risk of inducing renal failure
l Intolerance to antihypertensive medications image. The left renal usually arises obliquely from the aorta and
is best visualized with a 208 to 408 left anterior oblique (LAO)
Abbreviations: RAS, renal artery stenosis; ACEI, angiotensin convert- image (27). To accommodate these differences, initial angio-
ing enzyme inhibitors; ARB, angiotensin receptor blockers. graphic imaging is often performed in the 158 LAO position.
Review of prior CT or magnetic resonance (MR) angiographic
images will often allow precise identification of the optimal
pressure—is a better predictor of hemodynamic severity of
imaging angulation needed for stenting.
RAS than angiographic measurements (24).
The renal arteries can on occasion have a sharp caudal
angulation. In cases in which this precludes successful lesion
Anatomic Considerations for Renal Intervention crossing or results in inadequate guide support for stenting,
Several different variants of renal FMD have been described, alternative approaches such as a brachial or radial puncture can
the most common being medial fibroplasia (4). Since FMD be used (28). These techniques have been facilitated in recent
Figure 50.1 Fibromuscular dysplasia (FMD) of the renal arteries. (A) Abdominal aortogram showing bilateral FMD. There is a “string of
beads” appearance in the mid to distal portions of both renal arteries, more pronounced on the right. Also note the enlarged inferior
mesenteric artery due to FMD occlusion of the superior mesenteric artery (arrow). (B) Selective right renal arteriogram more clearly shows the
beaded appearance typical of medial fibroplasia. (C) After angioplasty, there is a smoother lumen in the treated portion of the renal artery.
Pressure measurements showed no residual translesional gradient.
[2/7/010/20:55:16] [625–639]
RENAL AND MESENTERIC ARTERY INTERVENTIONS 627
years by the routine use of low-profile stent systems that can be Queensbury, New York, U.S.) can often be useful. With this
placed through smaller sheath introducers. technique, the reverse catheter is formed using a floppy-tipped
The use of distal embolic protection devices (DEP) 0.035-in. guidewire either in the aorta below the renal arteries
imposes other anatomic considerations. CT studies have dem- or at the thoracic aortic arch. The catheter is initially positioned
onstrated a relatively short average length of the renal artery below the renal artery, with the tip facing the side to be treated.
from origin to its first bifurcation, particularly on the left side With approximately 5 mm of wire protruding from the catheter
(29). This has a considerable impact for DEP selection, since the tip, the catheter is gently advanced cranially. When the renal
distance of many filters from cone to base may leave insuffi- artery is engaged, there will be a visible “quick flip” of the wire.
cient space for stent positioning within the renal artery. Simi- The wire is then advanced into the artery while simultaneously
larly, determinations of arterial diameter in the DEP “landing retracting the reverse curve catheter across the lesion. The renal
zone” can exceed the diameter of the protection device, render- guide is then gently advanced and rotated until it engages the
ing it ineffective, or necessitating a partial (segmental) protec- renal ostium, and exchange is made through the catheter for a
tion strategy (29). The recent introduction of larger protection stiffer stenting wire. The catheter is then removed and angio-
profile shorter span filters may be more anatomically suited for plasty or stenting performed.
renal artery use (30). “Don’t touch” technique This technique minimizes
Multiple renal arteries occur in up to 25% of patients. In manipulation in the perirenal aorta, theoretically reducing the
patients with accessory RAS, the use of smaller profile devices risk of plaque or cholesterol embolization into the kidney
is preferred. Although not yet defined, drug-eluting stents may during lesion crossing. There are several different don’t touch
play a role for the treatment of arteries <5 mm in diameter due techniques. The most commonly described method is to
to the higher restenosis rate of these lesions. advance a “J” wire beyond the guide tip into the suprarenal
aorta. This pushes the guide tip away from the aortic wall,
Equipment and Techniques preventing plaque disruption. A second curved 0.014 in. floppy
Renal interventions are most commonly performed via the tip wire is then used to cross the RAS, before removing the
common femoral artery approach, using shaped guides J wire, allowing the guide to engage the renal artery ostium,
(advanced through a sheath) or guide sheaths (containing an and proceeding with the procedure.
inner introducer and integrated hemostatic valve). A variety of “Buddy wire” technique Although not a method for
guide shapes can be utilized, with common shapes including lesion crossing, the buddy wire technique represent a useful
the renal double curve, internal mammary, and multipurpose strategy to allow successful procedure completion in cases
designs, with the objective of having a sheath tip that is when the severity of the RAS impedes balloon or stent place-
directed toward and aligns with the angulation of the renal ment. After initial lesion crossing, the guide is aligned with the
artery origin. Patients are routinely anticoagulated prior to renal artery ostium, and a second short transition 0.014-in.
treatment with either heparin or bivalirudin. guidewire is passed into the renal artery. Balloon or stent
There are three critical steps involved in PTRA and PTRS: advancement is then performed over the stiffer wire, with the
lesion traversal, treatment selection decision making (including buddy wire serving to stabilize the guide and allow more
determination of transstenotic pressures), and balloon/stent directed translation of pushing forces across the artery without
positioning and deployment. recoiling the guide.
Lesion Traversal Selecting the Interventional Strategy

Crossing the lesion is often the most difficult component of Uncomplicated FMD (without associated dissection) is treated
percutaneous renal revascularization, particularly in cases of with PTRA first, whereas ostial ARAS is usually approached
severe stenosis or occlusion. The selected guide should have with an intent to stent. Truncal or nonostial ARAS may be
sufficient breadth in its secondary curve to abut the contrala- managed with an initial attempt at PTRA alone. In these cases,
teral aortic wall so that there is sufficient support for control- the plaque causing the stenosis is not in direct contiguity with
ling and advancing the renal crossing wire. For horizontally the aorta, and there is a resulting proximal “normal” renal
oriented renal arteries, the guide is initially positioned imme- artery segment. Balloons and stents are sized to the normal
diately at or just above the renal artery. A floppy-tipped wire diameter of the renal artery, either proximal to the stenosis or
with a gentle curve on the tip is then directed toward the beyond any area of poststenotic dilatation. The contralateral
angiographically demonstrated remaining renal artery lumen; renal artery may also be used for vessel sizing if necessary.
imaging is performed using small contrast injections through Renal artery diameters are usually 5 to 6 mm in women, and
the sheath sidearm. Gently rotating the guide and/or wire tip 6 to 7 mm in men.
may allow engagement of the true lumen when crossing eccen- Intervention is not warranted for a <50% RAS. For
tric lesions. With a properly shaped guide, primary lesion stenoses causing 60% to 80% narrowing, determination of the
crossing is usually performed using a 2-cm floppy-tipped transstenotic pressure gradient is recommended (13). After
0.014- or 0.018-in. guidewire. Care must be taken to avoid crossing the lesion, simultaneous dual-channel pressure meas-
excessive pushing if there is resistance to wire advancement, urements are obtained from the renal artery distal to the
and it is important that the wire remain visualized with its tip stenosis and from the aortic sheath. Renal pressures may be
in first- or second-order branches during all phases of the recorded either through a 4-Fr catheter or, more recently, by
procedure. Wires advanced too distally can result in cortical using a pressure-sensing wire (32). The latter has the advantage
perforation and retroperitoneal hemorrhage (31). After cross- of preventing obturation of the renal artery resulting in spe-
ing, the stiff portion of the guidewire should extend across the ciously low-pressure measurements. Although there is no
aortorenal angle to allow subsequent balloon and stent passage. absolute gradient that is hemodynamically significant, a trans-
For more caudally directed renal arteries, the use of a stenotic gradient exceeding 20 mmHg and a >10% peak aortic
reverse curve catheter (e.g., Sos Omni, AngioDynamics, systolic gradient [(aortic-renal)/aortic <0.90] are two criteria
[2/7/010/20:55:16] [625–639]
frequently utilized to indicate a stenosis warranting interven- restrictions in maximal capacity and ability for complete pro-
tion. Fractional flow reserve can also be obtained as a determi- tection. Advancement of the occlusion balloon into upper or
nation of stenosis severity prior to PTRS. lower pole segmental arteries for partial renal protection has
been described. While partial protection captures less measur-
Balloon and Stent Positioning able debris than complete protection, differences in renal func-
The selected balloon should be limited to just slightly longer tion preservation between the two approaches have not been
than the RAS treated. Dilatation more than 1 cm beyond the proven (41). Wire mesh filters have larger diameters, but gen-
stenosis should be avoided to prevent distal dissection. Balloon erally are more rigid and longer devices, such that positioning
inflation is performed over 20 to 30 seconds. After engaging the requires at least 2 cm of normal artery beyond the planned
stenosis with partial balloon inflation, the balloon catheter shaft distal end of a stent. If this is not possible, partial protection
should be advanced slightly forward so that the balloon strategies can be employed. To facilitate lesion crossing, a
assumes a more horizontal course at the renal ostium; sharp buddy wire technique can be advantageous.
downward angulation of the balloon at the aorta can result in
an aortic tear (33). For ostial ARAS, the shortest possible stent
that will completely cover the lesion and extend 1 to 3 mm into Outcomes After Renal Intervention
aorta should be used. When there is uncertainty as to the exact Patency and Reintervention
location of the ostium due to extensive aortic plaque, it is PTRA alone for patients with FMD is associated with very high
preferred to have complete coverage by assuring that a small patency rates and low rates of reintervention for recurrent
amount of stent projects into the aorta. Ostial locator devices in stenosis. In contrast, PTRA for ARAS results in frequent steno-
clinical development have the potential to allow precise stent sis recurrence; stent placement has considerably lower reste-
positioning in difficult cases (34). nosis and is thus widely accepted as the standard of care for
these lesions (42,43).
Distal Embolic Protection Restenosis after PTRS for ARAS occurs in 15% to 20% of
Cholesterol, platelet, and plaque embolization occurs in almost cases and is more frequent with initial stent diameters <5 to
all cases of renal intervention (35). In ex vivo models, embolic 6 mm (44,45). Renal stent fracture has also been recently asso-
debris has been recorded during each phase of the procedure, ciated with restenosis in approximately 20% of cases (46) and is
including wire traversal, balloon dilation, stent insertion, and not surprising given the tremendous respiratory motion of the
stent deployment (35). One recent study suggested a high kidney and resulting “fulcrum-like” stress on ostially located
preponderance of platelet-rich embolic debris and suggested stents (47). Not all patients with restenosis require target lesion
a role for a dual strategy of distal embolic protection and revascularization (TLR). In a large single-center experience by
intravenous glycoprotein IIB/IIIA receptor antagonism (36). Bates et al. of 748 patients, revascularization was performed in
Several different types of DEP have been used in the renal 10% of patients at nearly four-year follow-up after renal artery
artery, including wire mesh filters (37–39), fibrous mesh filters stenting (48). The initial use of sirolimus drug-eluting stents has
(30), and balloon occlusion devices (38,40). There are several been associated with a reduction in late lumen loss but non-
technical and anatomic considerations that are unique to renal substantial differences in binary restenosis or TLR (46).
DEPs (Table 50.2). Principal among technical issues are device The management of in-stent renal artery restenosis is
rigidity and crossing profile, which can affect the ability to uncertain. Strategies have included balloon angioplasty alone
position the DEP beyond the stenosis without injury, and (49), angioplasty with repeat stent placement (stent-in-stent
characteristics of the landing zone (position of filter position- angioplasty) (49,50), and covered stent placement (51). In a recent
ing), which may limit filter positioning or the ability for vessel series of 34 patients with injection site reaction (ISR), a stent-in-
wall apposition and complete protection. stent angioplasty was superior to repeat PTRA alone, with
With regard to device crossing as well as DEP length, recurrent ISR in 29% versus 71%, respectively (p ¼ 0.02) (50).
fibrous mesh filters and balloon occlusion devices have more
favorable characteristics, although occlusion balloons have Clinical Results
The results of PTRS for hypertension are shown in Table 50.3
(52–60). Revascularization in most series results in reproducible
Table 50.2 Unique Considerations for Distal Embolic Protection and sustainable reductions in both systolic and diastolic blood
Devices in the Renal Arteries pressure. In the largest published registry series to date, Dorros
et al. reported on 1058 successfully stented patients, including
Consideration Reason
901 (85%) who were treated for poorly controlled hypertension.
Length of renal arteries Right renal artery distance to Follow-up data demonstrated a durable and statistically signif-
bifurcation 43 15 mm icant improvement in blood pressure control (at 4 years: systolic,
Left renal artery distance to 168 þ 27 to 147 þ 21 mmHg; diastolic 84 þ 15 to 78 þ 12 mmHg;
bifurcation 33 14 mm
p < 0.05) as well as a significant decrease in the number of
Diameters of renal arteries Average “landing zone” diameter
5.8 mm antihypertensive medications (2.4 þ 1.1 to 1.8 þ 0.9 at 3 years;
Device crossing profile Severe stenosis p ¼ 0.05). Two recently published core laboratory adjudicated
Device rigidity RAS often tight and angulated industry-sponsored prospective trials support these findings. In
Filter pore size Cholesterol crystal 10–55 mm the ASPIRE-2 study, systolic/diastolic blood pressure was
Occlusive vs. filtration Risk of warm ischemia reduced from 168 25/82 13 mmHg at baseline to 149
devices 25/77 12 mmHg at two years (p < 0.001) in 208 patients (56).
Ease of use and versatility Anatomic variants such as proximal The number of antihypertensive medications was changed from
bifurcations common 2.8 0.9 at baseline to 2.3 1.3 at follow-up (p < 0.001). The
Abbreviation: RAS, renal artery stenosis. RENAISSANCE study evaluated 100 patients for up to three
[2/7/010/20:55:16] [625–639]
Table 50.3 Hypertension Outcomes After Renal Artery Stenting

Systolic BP (mmHg) Diastolic BP (mmHg)
Author N Follow-up (mo) Baseline Postprocedure Change Baseline Postprocedure Change Comments
Dangas, 2001 (52) 131 Mean 15 9 (n ¼ 118) 170 25 145 20 – 84 14 74 12 – 48% improved
Dorros, 2002 (53) 1058 12 168 27 146 24 – 84 15 75 12 –
Zeller, 2004 (60) 340 6 (n ¼ 278) 144 19 136 15 – 79 11 72 9 –
12 (n ¼ 211) 133 15 – 75 10 –
Sapoval, 2005 (58) 52 6 (n ¼ 46) 172 152 – 92 85 – 5% cure,
61% improved
Tsao, 2005 (59) 34 6 (n ¼ 33) 148 4 130 2 – 78 3 70 2 – 85% cure/improved
Rocha-Singh, 2005 (56) 208 9 (n ¼ 178) 168 25 149 24 – 82 13 77 12 –

Ruchin, 2007 (57) 89 Mean: 28 161.7 29.5 138.7 17.9 – 78.4 13.8 76.7 10.8 –
Goncalves, 2007 (54) 46 Median: 23 177 30 135 28 42.11 37.85 98 17 83 8 15 43% improved
Mean: 24.5 15.2
Rocha-Singh, 2008 (55) 100 9 (n ¼ 92) 156 18 148 22 9 24 75 12 75 9 –
36 141 22 16 29 71 9 4 15
Abbreviation: BP, blood pressure.
RENAL AND MESENTERIC ARTERY INTERVENTIONS
629
[2/7/010/20:55:16] [625–639]
Table 50.4 Renal Function Outcomes After Renal Artery Stenting

Mean serum creatinine (mg/dL) Categorical outcomes (%)a
First author, year n Follow-up (mo) Baseline Follow-up p Improved Stabilized
Dorros, 2002 (53) 1058 1.7 1.1 1.3 0.8 <0.05 – –
Rocha-Singh, 2002 (65) 51 2.3 0.9 1.7 0.7 <0.001 – –
Zeller, 2003 (21) 215 12 1.21 1.10 0.047 51 –
Zeller, 2004 (60) 354 34 – – – 10 39
Ramos, 2003 (63) 105 1.7 0.9 1.4 0.7 <0.0001 – –
Gill, 2003 (61) 100 25 – – – 31 42
Ilkay, 2004 (62) 13 2.6 0.9 1.8 0.6 <0.001 – –
Rivolta, 2005 (64) 52 24 2.9 1.8 – – 16 60
Rocha-Singh, 2008 63b 9 – – – 91
(ASPIRE-2 Study) (55)
Kashyap, 2007 (18) 125 12 2.2 0.9 2.4 1.5 0.10 42 25
Bates, 2008 (77) 111b 36 – – – 23 52
Rocha-Singh, 2008 100 36 1.3 0.4 1.4 0.6 0.1038 – –
(RENAISSANCE Trial) (55)
a
Definitions for improvement and stabilization vary by report.
b
Cohort of study patients with baseline serum creatinine 1.5 mg/dL.
years (55). Average systolic blood pressure was noted to be small platelet-rich emboli (36) and suggested an advantage for
reduced by 15 28 mmHg (p ¼ 0.0003), with diastolic pressure concomitant antiplatelet therapy and DEP use also seen by
lowered by 4 15 mmHg (p ¼ 0.0510). Edwards (86). In an evaluation of 63 patients with ischemic
The majority of patients undergoing RAST for renal nephropathy undergoing RAST with balloon occlusion embolic
insufficiency will have functional benefit, defined either as protection, Holden et al. noted stabilization or improvement of
stabilization or improvement in either serum creatinine or renal function after intervention in 97% of patients regardless of
calculated glomerular filtration rate (GFR) (Table 50.4) initial chronic kidney disease severity, although improvement
(18,22,53,55,56,60–65). Overall, between 50% and 90% of was more likely in patients with less initial functional impair-
patients will benefit. Studies evaluating single kidney or total ment (39). Of note, improvements in serum creatinine (SCr) were
GFR after RAST have consistently demonstrated incremental also more likely when visible debris was captured, although it is
positively directed changes when plotting the change in slope not clear whether this was due to technique or device failures or
of inverse serum creatinine curves prior to and following the occurrence of microemboli below the device filtration capac-
intervention (66–68). Predictors of benefit include an absence ity. Sanjay and colleagues recently reported on protected PTRS
of diabetes (69) or proteinuria (70), bilateral disease (71), recent using filter-type devices in 23 patients with baseline renal
rapid declines in renal function (18,72), and lower baseline insufficiency. In this group, the collective mean modification
serum creatinine (66,73,74). However, none of these is absolute, of diet in renal disease equation (89) estimated GFR rose from
and benefit may still be observed in patients with measurable 32.9 0.9 mL/min prior to intervention to 41.3 13.7 mL/min
proteinuria (75), diabetes (22,70,76), and advanced renal insuf- at last follow-up (8 5 months). K-DOQI stage (90) of chronic
ficiency (53,77). Recent reports have even described renal kidney disease improved in 26%, remained stable in 70%, and
recovery for dialysis-dependent patients after RAST (78). deteriorated in only 4% (86). In a series by Edwards et al.,
There are several reasons attributed for continued renal improvements in renal function were noted in 54% of 26 patients
decline in patients undergoing RAST for renal insufficiency. (32 lesions) with renal insufficiency (40). Randomized trials
These include progression of underlying intrinsic renal pathol- utilizing optimized renal protection devices will be necessary
ogy, immutable microcirculatory disturbance, contrast-induced to clearly establish the value for this approach.
nephropathy, and procedure-related embolization. A variety of
strategies should be considered for nephroprotection when per- Cardiovascular Events and Mortality
forming renal interventions in patients with baseline renal insuf- There are a multitude of pathophysiological alterations that
ficiency. Useful approaches include vigorous hydration (79), occur in RAS (91) beyond the classical Goldblatt model of
alkaline (bicarbonate sodium) infusions (80), use of diluted and hypertension (92). Neurohumoral activation results in diastolic
low-dose iso-osmolar nonionic contrast (81), CO2 angiography dysfunction, which in addition to aldosterone-mediated volume
(82), periprocedural administration of oral N-acetylcysteine (83) overload and angiotensin II-stimulated peripheral vasoconstric-
alone or in combination with bicarbonated saline infusion (84), tion results in left ventricular hypertrophy, cardiovascular
and the selective or routine use of distal embolic protection. amplification, diastolic dysfunction, and RAS-related cardiac
Several studies have evaluated outcomes of renal inter- disturbance syndromes including “flash” pulmonary edema
vention utilizing DEP, with most showing improved results and and precipitation of unstable angina. PTRS has been repeatedly
a lower rate of subsequent renal failure when compared with shown to mitigate the subsequent risk of CHF and pulmonary
historical controls (30,39,40,85). Capture of visible embolic edema in patients with RAS (93–95). In addition, at least one
debris occurs in 35% to 100% in reported series and appears to study has shown that revascularization of RAS in patients with
be more common with balloon occlusion techniques (Fig. 50.2) coronary artery disease and angina reduces anginal severity
(37,39,40,85–87). However, the majority of liberated embolic even without coronary revascularization (2).
particle may be <60 mm in size, and thus not macroscopically RAS is an independent risk factor for adverse cardiac
visible (88). The RESIST trial has demonstrated a propensity for events and cardiac mortality, with a graded risk correlation
[2/7/010/20:55:16] [625–639]
Figure 50.2 Treatment of atherosclerotic RAS using balloon occlusion distal embolic protection. (A) Digital subtraction abdominal
aortography demonstrates severe bilateral ostial RAS. (B) After selective catheterization and crossing of the left renal artery with a
GuardwireTM distal embolic protection device (Medtronic, Santa Rosa, California, U.S.), the balloon is inflated with resulting distal flow
occlusion. (C) The selective catheter is reduced and removed (over the occluded distal balloon), advancing the renal double curve guide to
the renal origin for stent positioning. Note the reflux of contrast into the aorta due to distal renal artery occlusion, facilitating stent positioning.
The stent was initially inserted too deeply into the renal artery and was subsequently retracted for final positioning. (D) Completion
angiography after stenting and deflation of the distal occlusion balloon shows a widely patent renal artery with preserved intrarenal flow.
Filtration of aspirated blood showed visible captured embolic debris. Total renal artery occlusion time was 8.5 minutes. Abbreviation: RAS,
renal artery stenosis.
with stenosis severity (96) and baseline renal impairment (97). nephropathy have a particularly grim prognosis, with an
Patients with RAS have higher mortality compared with annualized mortality rate in this cohort of almost 30% (99).
matched individuals without renovascular disease that is not Neurohumoral activation appears to lessen after PTRS.
clearly related to degree of hypertension or concomitant exis- Measurements of forearm vasoreactivity, serum endothelin,
tence of renal insufficiency (3). In a large study of 550 patients and left ventricular hypertrophy all improve after successful
screened for renovascular disease during peripheral angiogra- intervention (100–102). At least one retrospective study, con-
phy, the presence of RAS was an independent risk factor for ducted by Pizzolo et al., has demonstrated a survival advantage
subsequent mortality (98). Death over a 3.8-year period of for patients with RAS treated with endovascular versus non-
follow-up occurred in 27.6% of patients without RAS (n ¼ interventional therapy, with a mean 28-month cumulative
362), 54.3% of patients with RAS < 75% (n ¼ 94), and 71.4% probability of survival of 86.7% compared with 67.1% (101).
of patients with RAS 75% (n ¼ 35) (p ¼ 0.0001). Notably, Despite this, clear evidence of a survival benefit after RAST is
differences in survival were observed at all levels of renal lacking (10,103). Overall survival after RAST is to a large degree
insufficiency (98). Patients with dialysis-dependent ischemic determined by baseline renal impairment (71,104), although
[2/7/010/20:55:16] [625–639]
recent evidence suggests that an improvement in renal function However, abrupt hypotension is uncommon. Antihypertensive
in initially azotemic RAS patients undergoing RAST may therapy should otherwise follow established guidelines (111).
be associated with subsequent reduced mortality (105). The In patients with renal insufficiency, serum creatinine and GFR
RENNAISSANCE study provides a contemporary prospective should be measured within one week of intervention as well as
view of patients undergoing RAST, and found a three-year at other clinical follow-up appointments.
survival of almost 90% (55). Similar improved survival after Due to the recognized risk of restenosis after stenting,
renal revascularization has been demonstrated in both surgical duplex surveillance is recommended (55). Baseline studies
(106) and stent (107) trials. The NIH-sponsored CORAL trial is obtained within one to two weeks of intervention provide a
currently underway and will collect data to definitively evalu- comparison for subsequent studies performed at three to six
ate the occurrence of attributable cardiac and renal events in months, then annually for three to five years. Recurrent clinical
patients with RAS randomized to either optimal medical ther- events, particularly declining renal function or accelerated
apy alone or both medical therapy and RAST (7). hypertension, warrant prompt evaluation. Since duplex criteria
for restenosis within stents may be different than for de novo
RAS (112), particularly in patient with aortic endografts (113),
Complications of Renal Intervention
standards need to be developed and validated at local vascular
Renal intervention is generally safely performed. Risks of the
laboratories. In questionable cases, CT angiography may be
procedure are related to puncture site complications, contrast-
valuable for detecting in-stent restenosis (114).
induced nephropathy, aortic or renal embolization including
cholesterol embolization syndrome, and injury to the renal
artery during lesion crossing, angioplasty, or stent placement. Conclusions
General procedural risks include periprocedural cardiac events RAS is a common clinical problem. The safety of renal inter-
and mortality, although both of these are distinctly uncommon. ventions has improved with evolving techniques, and there is
Complications may be either clinically relevant (e.g., major strong albeit predominantly observational data supporting
complications) or procedurally occurring events without antici- clinical value for the treatment of associated hypertension,
pated clinical sequelae. Nonclinically important procedural renal failure, or pulmonary edema. Demonstration of a cardio-
events that increase the time or complexity of the procedure vascular survival benefit will be a critical determinant of long-
are termed “technical-procedural complications” (106) and term acceptability and is currently the subject of a large inter-
would include small groin hematomas not increasing the national multicenter prospective trial.
level of care, minor arterial dissections, and suboptimal stent
positioning or need for a second stent placement. This last
event is minimized by optimal profiling of ostial lesions before CHRONIC MESENTERIC ISCHEMIA
stent deployment. Introduction
The Society of Interventional Radiology has established Chronic mesenteric ischemia (CMI) or mesenteric angina is an
guidelines delineating thresholds for major complications (108). uncommon but not rare form of peripheral vascular disease.
On the basis of review of large meta-analyses, these established The disease usually develops after the age of 60 and is threefold
thresholds for 30-day mortality of 1%, periprocedural renal more common in women than men. Patients commonly have
failure or exacerbation of 2% to 5%, access site hematoma prior vascular disease, affecting the lower extremities, coronary
requiring intervention of 5%, and symptomatic embolization arteries, renal arteries, or cerebral arteries (115). Risk factors for
or procedure-related renal artery occlusion of approximately atherosclerosis, including cigarette smoking, hypertension, and
5% (108). Overall, major complications occur in approximately diabetes mellitus, are common. The classic symptomatic triad—
5% (109,110). As noted earlier, renal embolic events may be which typically occurs with advanced disease—is postprandial
minimized with the routine administration of antiplatelet pain, fear of eating, and involuntary weight loss. The charac-
agents and the use of distal embolic devices in selected cases. teristic pain is chronic and dull, begins 15 to 30 minutes after
meals, and persists for 1 to 4 hours thereafter (116). As the
disease progresses, the pain becomes progressively more
Postprocedural Care severe and longer lasting and occurs after eating smaller
All patients undergoing renal intervention should be main-
amounts of food.
tained on antiplatelet therapy for at least several months. While
Physical examination often reveals an abdominal bruit
the optimal antiplatelet regimen has not been established,
(117). Signs of weight loss and malnutrition, such as temporal
clopidogrel is the most commonly used agent and the de
wasting, are common. Many patients have evidence of periph-
facto standard of care. Some investigators apply the coronary
eral vascular disease, such as absent distal pulses and trophic
principle of dual antiplatelet therapy with aspirin and clopi-
changes in the feet, and of coronary artery disease, such as
dogrel for one month followed by long-term aspirin alone,
electrocardiographic abnormalities (118).
although this may be associated with a higher hemorrhagic
risk. No additional anticoagulation is necessary postprocedur-
ally. For patients with FMD, antihypertensive medication may Anatomic Considerations
be withheld following the procedure and the blood pressure The abdominal viscera are supplied by the celiac axis, superior
closely monitored to determine the need for reinstitution of mesenteric artery (SMA), and inferior mesenteric artery (IMA).
therapy. Since blood pressure in this scenario is frequently The occlusive disease progresses slowly in CMI permitting
renin mediated, angiotensin-converting enzyme inhibitors or collaterals to develop to prevent bowel infarction. The celiac
angiotensin II receptor blockers are preferred. For patients with artery and SMA communicate by anastomoses between the
ARAS, blood pressure medications should be considered, superior pancreaticoduodenal branch of the gastroduodenal
although the most recently added or increased drug may artery, which arises from the hepatic artery, and the inferior
sometimes be held pending an evaluation of clinical results. pancreaticoduodenal artery, which arises from the first part of
[2/7/010/20:55:16] [625–639]
exchanged for a selective catheter. Our preference is to use a

reverse curve catheter during femoral access and an angled
multipurpose catheter during a brachial approach (Fig. 50.4).
The catheter is navigated into the abdominal aorta and engages
the target vessel, either celiac or SMA. If not already given,
heparin (usually 5000 units) should be administered at that
point. The lesion can then be crossed with a 0.035-in. guidewire.
Alternatively, a 0.014-in. wire can be used if one plans to use a
rapid exchange system. A 6- or 7-Fr guiding catheter or sheath is
then placed just proximal to the lesion, and the balloon and/or
stent is delivered to the lesion. Angiography through the guiding
system is obtained to pinpoint the stenosis, and the angioplasty
or stent placement follows. Alternatively, a no-touch technique
has been described and can be used if the origin of the target
vessel is very calcified or ulcerated. With this technique, the
guide catheter is positioned in the aorta and is maintained
straight by means of a 0.035-in. wire that keeps the catheter
from engaging the takeoff of the target vessel. A 0.014-in. wire is
then inserted in the guiding catheter, the 0.035-in. wire is
Figure 50.3 Diagram of arterial supply to the viscera demonstrates
removed, and the lesion is crossed with the finer 0.014-in. wire
the intrinsic collateral pathways seen in chronic mesenteric ische-
mia. The major collateral pathway between the celiac and superior (125). The guiding catheter can at this point be advanced to
mesenteric artery is the anastomosis between the superior and engage the lesion and the intervention follows. After the proce-
inferior pancreaticoduodenal arteries (thin arrow). The major path- dure and if a stent has been deployed, the patient stays on
way between the superior and inferior mesenteric arteries is the clopidogrel and aspirin for at least six months.
marginal artery of Drummond (thick arrow). The question of primary versus selective stenting has not
been resolved. It appears to be a general consensus that an
unsatisfactory result after angioplasty alone as evidenced by
residual stenosis >30% or a >15-mmHg pressure gradient
the SMA (Fig. 50.3). When the celiac artery is stenosed or across the lesion, calcified ostial or high-grade eccentric sten-
occluded, blood flows from the SMA to branches of the celiac oses, chronic occlusions, or the presence of dissection after
artery, and when the proximal SMA is occluded or stenosed, angioplasty all constitute indications for stent placement. Bal-
blood flows in the reverse direction. An anastomosis may also loon-expandable stents that offer precise placement are favored
occur between the ascending division of the left colic branch of by most authors. Placement with 1- to 2-mm protrusion into the
the IMA and the left division of the middle colic branch of the aortic lumen is advised. Postdilation is performed, and depend-
SMA in response to stenosis or occlusion of either the IMA or ing on the angiographic result, the pressure gradient across the
proximal SMA. This anastomosis is called the marginal artery or lesion may be measured.
arc of Riolan (119). Occasionally, the IMA, which is usually
small, may become enlarged and supply the entire abdominal
viscera (Fig. 50.1) (120,121). Because of this rich collateral net- Complications
work, at least two of the three major mesenteric vessels must be Access complications are by far the most common and can take
diseased before symptoms of CMI occur (122). the form of either hemorrhage or thrombosis. Hemorrhage is
common in femoral and high brachial or axillary approaches.
Bleeding into the axillary sheath has the potential to perma-
Indications
nently compromise nerve function; therefore, early diagnosis
The goals of revascularization in patients with symptomatic
and prompt evacuation are essential to minimize morbidity.
CMI are to improve symptoms and nutritional status and pre-
Careful technique that involves an ultrasound-guided stick
vent intestinal infarction (123). Prophylactic mesenteric revascu-
using a micropuncture needle is of paramount importance to
larization is rarely performed in the asymptomatic patient
prevent this complication. Thrombosis occurs almost exclu-
undergoing an aortic procedure for other indications (124).
sively in the brachial artery that is small in size and can be
totally occluded by interventional sheaths. Rapid hepariniza-
Fundamentals and Equipment tion after sheath insertion is usually an adequate preventive
Both the brachial and femoral approaches have been utilized in measure. The status of the radial and ulnar circulation should
mesenteric angioplasty and stenting. The former appears to be documented preoperatively, and any evidence of compro-
have a slight advantage in cases of acute angulation at the mise after the completion of the endovascular procedure
mesenteric vessel origin off the aorta. A 6-Fr sheath is inserted should be aggressively treated with thromboembolectomy of
via the access vessel. If the brachial artery has been accessed, the brachial artery to avoid permanent ischemic sequelae.
then early heparinization is advised to prevent thrombosis at Entering a subintimal plane while trying to cross a high-
the sheath insertion site. A pigtail catheter is advanced through grade stenosis or occlusion causes dissection, which necessi-
the sheath into the abdominal aorta, and a diagnostic arterio- tates stent placement to avoid distal propagation and arterial
gram is performed, first in the anterior-posterior direction to occlusion. If reentry to the true lumen and successful wire
assess the SMA and celiac branches, and then in the lateral advancement through the lesion are not possible, then conver-
projection to assess the arterial orifices and confirm the pression to an open operation may be necessary. Bowel ischemia is
ence and severity of stenosis. The pigtail is subsequently a related but infrequent complication that develops in cases of
[2/7/010/20:55:16] [625–639]
Figure 50.4 Chronic mesenteric ischemia: 65-year-old female with 5 months of progressive postprandial abdominal pain. The patient has
lost 35 lbs over that time period. CT and conventional angiograms demonstrate occlusion of the infrarenal aorta (bold arrow, A) and high-
grade atherosclerotic stenosis of the celiac and superior mesenteric arteries (white arrows, B). The inferior mesenteric artery is occluded at its
origin secondary to the aortic occlusion. From a brachial approach, a multipurpose guide catheter is advanced to the visceral aorta and used
to deliver balloon-expandable stents into the ostia of both the superior mesenteric (C,D) and celiac (E,F) arteries with excellent angiographic
results. The patient remains symptom-free at 24-month follow-up. Abbreviation: CT, computed tomography.
underestimated dissection after percutaneous transluminal 60% (Table 50.5) (123–125,132–135). This fact has led to our
angioplasty or as a result of distal embolization after crossing institution initiating an aggressive duplex surveillance protocol
long occlusions. If intestinal malperfusion is suspected intra- following mesenteric revascularization procedures to improve
operatively and confirmed angiographically, then standard the primary patency of these procedures.
catheter-based salvage techniques are available and should be
implemented immediately. When these are not successful, or if
abdominal symptoms from presumed intestinal ischemia have Special Considerations
been established, abdominal exploration with bowel inspection Acute Mesenteric Ischemia
and thromboembolectomy is indicated. Other complications The role of endovascular therapy in acute mesenteric ischemia
that are common to all the endovascular procedures, such as (AMI) remains controversial. In cases of AMI, assessment of
renal failure and anaphylactic reaction, may also infrequently intestinal viability is crucial and can be achieved only with
occur. One recognized pitfall of endovascular therapy of CMI is abdominal exploration and direct bowel inspection. That point
a relatively high restenosis and reintervention rate of 15% to aside, it is worth mentioning that endovascular treatment is an
[2/7/010/20:55:16] [625–639]
Table 50.5 Outcomes of Recent Series of Endovascular Treatment for Chronic Mesenteric Ischemia
Success 30-Day No. of patients/
Author, year N (technical/clinical) Complications mortality Restenosis mean follow-up Reintervention
Lim, 2005 (131) 8 7/7 2 0 2 8/52 2
Landis, 2005 (134) 29 28/26 4 2 10 29/28 10
AbuRahma, 2003 (125) 22 21/21 0 0 7 18/26 5
Matsumoto, 2002 (135) 33 27/24 5 0 5 29/38 6
Kasirajan, 2001 (124) 28 28/17 5 3 5 19/24 19
Nyman, 1998 (133) 5 4/4 2 0 3 5/24 3
Rose, 1995 (123) 8 3/4 2 1 2 6/9 1
Totals 133 8/83 20 6 34 114/28 46
appropriate alternative for the occasional patient who is a Comparison of Endovascular with Open Surgical Technique
prohibitive operative risk and does not have frank peritoneal Surgical intervention for mesenteric occlusive disease has tra-
signs on physical examination as well as for those patients who ditionally been the treatment of choice. The number of vessels
have a contaminated peritoneal cavity and lack an autogenous revascularized has often been reported to influence the long-
conduit for the performance of a mesenteric or celiac bypass. term outcome. However, it has been noted that revasculariza-
Along these lines, Boley et al. (126) used arteriography and tion of the SMA is of paramount importance and provides
subsequent catheter-based interventions or transcatheter vaso- optimal long-term symptomatic relief, even if revascularization
dilator therapy as the initial or sole therapy of AMI. Catheter- of other compromised arteries is not possible (136,137). Open
directed thrombolysis (127,128), as well as percutaneous angio- surgical techniques (SMA endarterectomy or aortomesenteric
plasty (129), has also been used with good results in the or celiac bypass grafting) have achieved an immediate clinical
treatment of acute mesenteric embolism. Demirpolat et al. success that approaches 100%, surgical mortality rates from 0%
(130) reported three patients with increasing abdominal pain to 17%, and operative morbidity rates that range from 19% to
but without peritoneal findings who were treated percutane- 54% in a number of different series (115,138–140). The median
ously with good outcome. Lastly, Lim and colleagues (131,132) hospital stay has been reported to be 14 days (140). Recently,
have reported the use of an endovascular approach in surgi- two reports of surgical revascularization for CMI revealed five-
cally unfit patients with good result. All these authors empha- year survival rates of 61% to 64% and a nine-year assisted
sized the importance of performing endovascular treatment in primary graft patency rate of 79% (138,139). However, overall
the absence of peritoneal signs and only when bowel viability patency rates as high as 93% have been reported (134).
can be assessed either clinically or with imaging techniques. In comparison to the above historical results and based
on the data from a recent meta-analysis (141), mesenteric
Total Occlusions angioplasty and stenting demonstrate slightly inferior technical
The presence of total occlusion implies a technically more and clinical success rates (Table 50.5). Long-term patency rates
demanding procedure, but it does not constitute a contraindi- appear to also be superior with the open technique. There is a
cation for endovascular intervention. Only a few attempts to general consensus, however, that the endovascular approach is
cross occlusions were seen in the early studies, and Kasirajan associated with lower morbidity and mortality rates. Histori-
et al. (124) noted that it was more common for their group to cally, open surgical repair of chronic mesenteric occlusion is
use an open procedure to treat patients with occluded vessels. associated with mean morbidity of 29% (range 19–54%) and a
A major concern at the time was the potential for plaque mean 30-day mortality of 7% (range 0–17%) (125). There is no
fragmentation, with subsequent distal embolization. This, how- randomized study comparing open versus endovascular inter-
ever, has not been confirmed by authors who crossed and vention, and given the rarity of this condition, structuring such
treated occluded mesenteric vessels (133). The length of the a study would be a challenging task. Current consensus is to
occlusion correlates with plaque burden and seems to be an reserve endovascular intervention for the high-risk and older
important predictor of the likelihood for distal embolization; patients or for the individual patient with unclear symptoma-
however, given the few patients studied, this has not been tology and questionable diagnosis.
statistically confirmed. Low-profile systems and evolving
expertise have now made successful treatment of occluded Median Arcuate Ligament Syndrome
vessels feasible; Landis et al. (134) treated nine patients who The median arcuate ligament (MAL) syndrome results from
had mesenteric occlusion and reported 100% technical success, compression of the celiac artery by the MAL. Compression by
clinical success, and primary one-year patency. the adjacent sympathetic plexus may also contribute to the
As a principle, treatment of SMA lesions takes priority celiac axis compression. Most patients are asymptomatic.
over treatment of lesions in the celiac axis or the IMA. Even this Symptoms, when present, mimic the clinical picture of CMI.
rule, however, has an exception. In a study by Matsumoto et al. Because of the extrinsic nature of the lesion, angioplasty or
(135), four patients with occluded SMA had only their IMAs stenting in these patients is associated with high failure rates
treated and remained asymptomatic in long-term follow-up. A and is not recommended by most authors (133,134). The typical
patent meandering artery can assure collateral circulation in the angiographic finding is that of a nonostial, eccentric, anterior,
event of a difficult-to-treat SMA occlusion. The feeding vessel and superior lesion of the celiac artery that becomes more
of such important collateral should be visualized in detail, and pronounced during deep expiration (Fig. 50.5) (142). The endo-
any stenotic lesions found should be treated aggressively. vascular approach can be of benefit in the occasional patient
[2/7/010/20:55:16] [625–639]
6. Edwards MS, Craven TE, Burke GL, et al. Renovascular disease

and the risk of adverse coronary events in the elderly: a pro-
spective, population-based study. Arch Intern Med 2005;
165:207–213.
7. Cooper CJ, Murphy TP, Matsumoto A, et al. Stent revasculariza-
tion for the prevention of cardiovascular and renal events among
patients with renal artery stenosis and systolic hypertension:
rationale and design of the CORAL trial. Am Heart J 2006;
152:59–66.
8. Palmaz JC, Kopp DT, Hayashi H, et al. Normal and stenotic renal
arteries: experimental balloon-expandable intraluminal stenting.
Radiology 1987; 164:705–708.
9. van Jaarsveld BC, Krijnen P, Pieterman H, et al. The effect of
balloon angioplasty on hypertension in atherosclerotic renal-
artery stenosis. Dutch Renal Artery Stenosis Intervention Coop-
erative Study Group. N Engl J Med 2000; 342:1007–1014.
10. Balk E, Raman G, Chung M, et al. Effectiveness of management
strategies for renal artery stenosis: a systematic review. Ann
Intern Med 2006; 145:901–912.
11. White CJ, Jaff MR, Haskal ZJ, et al. Indications for renal arte-
riography at the time of coronary arteriography: a science advi-
sory from the American Heart Association Committee on
Diagnostic and Interventional Cardiac Catheterization, Council
on Clinical Cardiology, and the Councils on Cardiovascular
Radiology and Intervention and on Kidney in Cardiovascular
Figure 50.5 Median arcuate ligament. Images from a lateral aorto- Disease. Circulation 2006; 114:1892–1895.
gram in a patient with symptoms of chronic mesenteric ischemia. At 12. Hirsch AT, Haskal ZJ, Hertzer NR, et al. ACC/AHA 2005 Prac-
expiration, there is an extrinsic compression on the superior wall of tice Guidelines for the management of patients with peripheral
the celiac artery (small arrow) from the anterior crus of the dia- arterial disease (lower extremity, renal, mesenteric, and abdomi-
phragm. At inspiration, the compression is relieved (large arrow). nal aortic): a collaborative report from the American Association
There is moderate atherosclerotic plaque in this vessel as well as for Vascular Surgery/Society for Vascular Surgery, Society for
the superior mesenteric artery. Cardiovascular Angiography and Interventions, Society for Vas-
cular Medicine and Biology, Society of Interventional Radiology,
and the ACC/AHA Task Force on Practice Guidelines (Writing
with unclear symptomatology for whom relief of symptoms Committee to Develop Guidelines for the Management of
after percutaneous transluminal angiography or stenting pre- Patients With Peripheral Arterial Disease): endorsed by the
dicts a good response to operative intervention and for the American Association of Cardiovascular and Pulmonary Reha-
high-risk, malnourished patient who will not be able to tolerate bilitation; National Heart, Lung, and Blood Institute; Society for
Vascular Nursing; TransAtlantic Inter-Society Consensus; and
an open surgical procedure.
Vascular Disease Foundation. Circulation 2006; 113:e463–e654.
13. Rundback JH, Sacks D, Kent KC, et al. Guidelines for the
Conclusion reporting of renal artery revascularization in clinical trials. Amer-
Endovascular management of patients with CMI offers a num- ican Heart Association. Circulation 2002; 106:1572–1585.
ber of significant advantages to the traditional open techniques 14. Mistry S, Ives N, Harding J, et al. Angioplasty and Stent for renal
of revascularization. It is associated with shorter hospital stay artery lesions (ASTRAL trial): rationale, methods and results so
far. J Hum Hypertens 2007; 21:511–515.
and low morbidity and mortality rates. Perioperative use of
15. De Bruyne B, Manoharan G, Pijls NH, et al. Assessment of renal
heparin, postoperative administration of antiplatelet agents, artery stenosis severity by pressure gradient measurements.
more liberal use of stents, improvement in the skill level, and J Am Coll Cardiol 2006; 48:1851–1855.
more careful patient selection, with increased awareness of the 16. Dubel GJ, Murphy TP. The role of percutaneous revasculariza-
MAL syndrome, all have the potential to improve the technical tion for renal artery stenosis. Vasc Med 2008; 13:141–156.
and short- and long-term clinical success rates. 17. Giroux MF, Soulez G, Therasse E, et al. Percutaneous revascula-
rization of the renal arteries: predictors of outcome. J Vasc Interv
Radiol 2000; 11:713–720.
REFERENCES 18. Kashyap VS, Sepulveda RN, Bena JF, et al. The management of
1. Schwartz CJ, White TA. Stenosis of renal artery: an unselected renal artery atherosclerosis for renal salvage: does stenting help?
necropsy study. Br Med J 1964; 2:1415–1421. J Vasc Surg 2007; 45:101–108; discussion 8–9.
2. Conlon PJ, Little MA, Pieper K, et al. Severity of renal vascular 19. Webster J, Marshall F, Abdalla M, et al. Randomised comparison
disease predicts mortality in patients undergoing coronary of percutaneous angioplasty vs. continued medical therapy for
angiography. Kidney Int 2001; 60:1490–1497. hypertensive patients with atheromatous renal artery stenosis.
3. Mui KW, Sleeswijk M, van den Hout H, et al. Incidental renal Scottish and Newcastle Renal Artery Stenosis Collaborative
artery stenosis is an independent predictor of mortality in Group. J Hum Hypertens 1998; 12:329–335.
patients with peripheral vascular disease. J Am Soc Nephrol 20. Radermacher J, Chavan A, Bleck J, et al. Use of Doppler ultra-
2006; 17:2069–2074. sonography to predict the outcome of therapy for renal-artery
4. Olin JW, Pierce M. Contemporary management of fibromuscular stenosis. N Engl J Med 2001; 344:410–417.
dysplasia. Curr Opin Cardiol 2008; 23:527–536. 21. Zeller T, Frank U, Muller C, et al. Predictors of improved renal
5. Murphy TP, Rundback JH, Cooper C, et al. Chronic renal function after percutaneous stent-supported angioplasty of
ischemia: implications for cardiovascular disease risk. J Vasc severe atherosclerotic ostial renal artery stenosis. Circulation
Interv Radiol 2002; 13:1187–1198. 2003; 108:2244–2249.
[2/7/010/20:55:16] [625–639]
22. Zeller T, Muller C, Frank U, et al. Stent angioplasty of severe 44. Bates MC, Rashid M, Campbell JE, et al. Factors influencing the
atherosclerotic ostial renal artery stenosis in patients with diabe- need for target vessel revascularization after renal artery stent-
tes mellitus and nephrosclerosis. Catheter Cardiovasc Interv ing. J Endovasc Ther 2006; 13:569–577.
2003; 58:510–515. 45. Lim ST, Rosenfield K. Renal artery stent placement: indications
23. Silva JA, Chan AW, White CJ, et al. Elevated brain natriuretic and results. Curr Interv Cardiol Rep 2000; 2:130–139.
peptide predicts blood pressure response after stent revasculari- 46. Nallusamy A, Sundaram V, Abraham G, et al. Recurrent hyper-
zation in patients with renal artery stenosis. Circulation 2005; tension from stent fracture. Kidney Int 2008; 73:1446.
111:328–333. 47. Draney MT, Zarins CK, Taylor CA. Three-dimensional analysis
24. Subramanian R, White CJ, Rosenfield K, et al. Renal fractional of renal artery bending motion during respiration. J Endovasc
flow reserve: a hemodynamic evaluation of moderate renal Ther 2005; 12:380–386.
artery stenoses. Catheter Cardiovasc Interv 2005; 64:480–486. 48. Zahringer M, Sapoval M, Pattynama PM, et al. Sirolimus-eluting
25. Kim HJ, Do YS, Shin SW, et al. Percutaneous transluminal versus bare-metal low-profile stent for renal artery treatment
angioplasty of renal artery fibromuscular dysplasia: mid-term (GREAT Trial): angiographic follow-up after 6 months and clin-
results. Korean J Radiol 2008; 9:38–44. ical outcome up to 2 years. J Endovasc Ther 2007; 14:460–468.
26. Leertouwer TC, Gussenhoven EJ, Bosch JL, et al. Stent placement 49. Zeller T, Rastan A, Schwarzwalder U, et al. Treatment of in-stent
for renal arterial stenosis: where do we stand? A meta-analysis. restenosis following stent-supported renal artery angioplasty.
Radiology 2000; 216:78–85. Catheter Cardiovasc Interv 2007; 70:454–459.
27. Beregi JP, Mauroy B, Willoteaux S, et al. Anatomic variation in 50. N’Dandu ZM, Badawi RA, White CJ, et al. Optimal treatment of
the origin of the main renal arteries: spiral CTA evaluation. Eur renal artery in-stent restenosis: repeat stent placement versus
Radiol 1999; 9:1330–1334. angioplasty alone. Catheter Cardiovasc Interv 2008; 71:701–705.
28. Kessel DO, Robertson I, Taylor EJ, et al. Renal stenting from the 51. Yan BP, Kiernan TJ, Rosenfield K. Covered stent for the treat-
radial artery: a novel approach. Cardiovasc Intervent Radiol ment of recurrent bilateral renal artery in-stent restenosis.
2003; 26:146–149. J Invasive Cardiol 2008; 20:E288–E292.
29. Thatipelli MR, Sabater EA, Bjarnason H, et al. CT angiography of 52. Dangas G, Laird JR Jr., Mehran R, et al. Intravascular ultrasound-
renal artery anatomy for evaluating embolic protection devices. guided renal artery stenting. J Endovasc Ther 2001; 8:238–247.
J Vasc Interv Radiol 2007; 18:842–846. 53. Dorros G, Jaff M, Mathiak L, et al. Multicenter Palmaz stent renal
30. Henry M, Henry I, Polydorou A, et al. Embolic protection for artery stenosis revascularization registry report: four-year fol-
renal artery stenting. J Cardiovasc Surg (Torino) 2008; 49:571–589. low-up of 1,058 successful patients. Catheter Cardiovasc Interv
31. Olteanu B, Olteanu C, Borelli C, et al. Embolization of a perfo- 2002; 55:182–188.
ration of a cortical renal artery occurring during percutaneous 54. Goncalves JA, Amorim JE, Soares Neto MM, et al. Clinical
renal angioplasty. Eur Radiol 2000; 10:1357. efficacy of percutaneous renal revascularization with stent place-
32. Colyer WR Jr., Cooper CJ, Burket MW, et al. Utility of a 0.014@ ment in atherosclerotic renovascular disease. Arq Bras Cardiol
pressure-sensing guidewire to assess renal artery translesional 2007; 88:85–90.
systolic pressure gradients. Catheter Cardiovasc Interv 2003; 55. Rocha-Singh K, Jaff MR, Lynne Kelley E. Renal artery stenting
59:372–377. with noninvasive duplex ultrasound follow-up: 3-year results
33. Rasmus M, Huegli R, Jacob AL, et al. Extensive iatrogenic aortic from the RENAISSANCE renal stent trial. Catheter Cardiovasc
dissection during renal angioplasty: successful treatment with a Interv 2008; 72:853–862.
covered stent-graft. Cardiovasc Intervent Radiol 2007; 30:497–500. 56. Rocha-Singh K, Jaff MR, Rosenfield K. Evaluation of the safety
34. Fischell TA, Malhotra S, Khan S. A new ostial stent positioning and effectiveness of renal artery stenting after unsuccessful bal-
system (Ostial Pro) for the accurate placement of stents to treat loon angioplasty: the ASPIRE-2 study. J Am Coll Cardiol 2005;
aorto-ostial lesions. Catheter Cardiovasc Interv 2008; 71:353–357. 46:776–783.
35. Hiramoto J, Hansen KJ, Pan XM, et al. Atheroemboli during renal 57. Ruchin PE, Baron DW, Wilson SH, et al. Long-term follow-up of
artery angioplasty: an ex vivo study. J Vasc Surg 2005; 41: renal artery stenting in an Australian population. Heart Lung
1026–1030. Circ 2007; 16:79–84.
36. Cooper CJ, Haller ST, Colyer W, et al. Embolic protection and 58. Sapoval M, Zahringer M, Pattynama P, et al. Low-profile stent
platelet inhibition during renal artery stenting. Circulation 2008; system for treatment of atherosclerotic renal artery stenosis: the
117:2752–2760. GREAT trial. J Vasc Interv Radiol 2005; 16:1195–1202.
37. Hagspiel KD, Stone JR, Leung DA. Renal angioplasty and stent 59. Tsao CR, Lee WL, Liu TJ, et al. Delicate percutaneous renal artery
placement with distal protection: preliminary experience with stenting minimizes postoperative renal injury and protects kid-
the FilterWire EX. J Vasc Interv Radiol 2005; 16:125–131. ney in patients with severe atherosclerotic renal artery stenosis
38. Henry M, Henry I, Klonaris C, et al. Renal angioplasty and and impaired renal function. Int Heart J 2005; 46:1061–1072.
stenting under protection: the way for the future? Catheter 60. Zeller T, Frank U, Muller C, et al. Stent-supported angioplasty of
Cardiovasc Interv 2003; 60:299–312. severe atherosclerotic renal artery stenosis preserves renal func-
39. Holden A, Hill A, Jaff MR, et al. Renal artery stent revasculari- tion and improves blood pressure control: long-term results from
zation with embolic protection in patients with ischemic nephr- a prospective registry of 456 lesions. J Endovasc Ther 2004; 11:
opathy. Kidney Int 2006; 70:948–955. 95–106.
40. Edwards MS, Craven BL, Stafford J, et al. Distal embolic protec- 61. Gill KS, Fowler RC. Atherosclerotic renal arterial stenosis: clinical
tion during renal artery angioplasty and stenting. J Vasc Surg outcomes of stent placement for hypertension and renal failure.
2006; 44:128–135. Radiology 2003; 226:821–826.
41. Kanjwal K, Haller S, Steffes M, et al. Complete versus partial 62. Ilkay E, Gunal IA, Yavuzkir M, et al. Effect of renal artery
distal embolic protection during renal artery stenting. Catheter stenting on renal function in patients with ischemic nephropathy.
Cardiovasc Interv 2009; 73:725–730. Jpn Heart J 2004; 45:637–645.
42. Baumgartner I, von Aesch K, Do DD, et al. Stent placement in 63. Ramos F, Kotliar C, Alvarez D, et al. Renal function and outcome
ostial and nonostial atherosclerotic renal arterial stenoses: a proof PTRA and stenting for atherosclerotic renal artery stenosis.
spective follow-up study. Radiology 2000; 216:498–505. Kidney Int 2003; 63:276–282.
43. van de Ven PJ, Kaatee R, Beutler JJ, et al. Arterial stenting and 64. Rivolta R, Bazzi C, Stradiotti P, et al. Stenting of renal artery
balloon angioplasty in ostial atherosclerotic renovascular disease: stenosis: is it beneficial in chronic renal failure? J Nephrol 2005;
a randomised trial. Lancet 1999; 353:282–286. 18:749–754.
[2/7/010/20:55:16] [625–639]
65. Rocha-Singh KJ, Ahuja RK, Sung CH, et al. Long-term renal 85. Holden A, Hill A. Renal angioplasty and stenting with distal
function preservation after renal artery stenting in patients with protection of the main renal artery in ischemic nephropathy:
progressive ischemic nephropathy. Catheter Cardiovasc Interv early experience. J Vasc Surg 2003; 38:962–968.
2002; 57:135–141. 86. Edwards MS, Corriere MA, Craven TE, et al. Atheroembolism
66. Harden PN, MacLeod MJ, Rodger RS, et al. Effect of renal-artery during percutaneous renal artery revascularization. J Vasc Surg
stenting on progression of renovascular renal failure. Lancet 2007; 46:55–61.
1997; 349:1133–1136. 87. Misra S, Gomes MT, Mathew V, et al. Embolic protection devices
67. Pattynama PM, Becker GJ, Brown J, et al. Percutaneous angio- in patients with renal artery stenosis with chronic renal insuffi-
plasty for atherosclerotic renal artery disease: effect on renal ciency: a clinical study. J Vasc Interv Radiol 2008; 19:1639–1645.
function in azotemic patients. Cardiovasc Intervent Radiol 1994; 88. Levey AS, Bosch JP, Lewis JB, et al. A more accurate method to
17:143–146. estimate glomerular filtration rate from serum creatinine: a new
68. Watson PS, Hadjipetrou P, Cox SV, et al. Effect of renal artery prediction equation. Modification of Diet in Renal Disease Study
stenting on renal function and size in patients with atheroscler- Group. Ann Intern Med 1999; 130:461–470.
otic renovascular disease. Circulation 2000; 102:1671–1677. 89. National Kidney Foundation. K/DOQI clinical practice guide-
69. Safian RD, Textor SC. Renal-artery stenosis. N Engl J Med 2001; lines for chronic kidney disease: evaluation, classification, and
344:431–442. stratification. Am J Kidney Dis 2002; 39:S1–S266.
70. Halimi JM, Ribstein J, Du Cailar G, et al. Albuminuria predicts 90. Rundback JH, Murphy TP, Cooper C, et al. Chronic renal
renal functional outcome after intervention in atheromatous ischemia: pathophysiologic mechanisms of cardiovascular and
renovascular disease. J Hypertens 1995; 13:1335–1342. renal disease. J Vasc Interv Radiol 2002; 13:1085–1092.
71. Dorros G, Jaff M, Mathiak L, et al. Four-year follow-up of 91. Goldblatt H, Lynch J, Hanzal RF, et al. Studies on experimental
Palmaz-Schatz stent revascularization as treatment for athero- hypertension. I. The production of persistent elevation of systolic
sclerotic renal artery stenosis. Circulation 1998; 98:642–647. blood pressure by means of renal ischemia. J Exp Med 1934;
72. Muray S, Martin M, Amoedo ML, et al. Rapid decline in renal 59:347–380.
function reflects reversibility and predicts the outcome after 92. Bloch MJ, Trost DW, Pickering TG, et al. Prevention of recurrent
angioplasty in renal artery stenosis. Am J Kidney Dis 2002; pulmonary edema in patients with bilateral renovascular disease
39:60–66. through renal artery stent placement. Am J Hypertens 1999; 12:1–7.
73. Beutler JJ, Van Ampting JM, Van De Ven PJ, et al. Long-term 93. Gray BH, Olin JW, Childs MB, et al. Clinical benefit of renal
effects of arterial stenting on kidney function for patients with artery angioplasty with stenting for the control of recurrent and
ostial atherosclerotic renal artery stenosis and renal insufficiency. refractory congestive heart failure. Vasc Med 2002; 7:275–279.
J Am Soc Nephrol 2001; 12:1475–1481. 94. Khosla S, White CJ, Collins TJ, et al. Effects of renal artery stent
74. Rundback JH, Manoni T, Rozenblit GN, et al. Balloon angioplasty implantation in patients with renovascular hypertension present-
or stent placement in patients with azotemic renovascular dis- ing with unstable angina or congestive heart failure. Am J Cardiol
ease: a retrospective comparison of clinical outcomes. Heart Dis 1997; 80:363–366.
1999; 1:121–125. 95. Messina LM, Zelenock GB, Yao KA, et al. Renal revascularization
75. Bali HK, Jha V. Nephrotic syndrome and recurrent pulmonary for recurrent pulmonary edema in patients with poorly con-
oedema in bilateral atherosclerotic renal artery stenosis: resolu- trolled hypertension and renal insufficiency: a distinct subgroup
tion following renal angioplasty and stenting. Natl Med J India of patients with arteriosclerotic renal artery occlusive disease.
2006; 19:253–254. J Vasc Surg 1992; 15:73–80; discussion 80–82.
76. Zuccala A, Losinno F, Zucchelli A, et al. Renovascular disease in 96. Bates MC, Campbell JE, Stone PA, et al. Factors affecting long-
diabetes mellitus: treatment by percutaneous transluminal renal term survival following renal artery stenting. Catheter Cardio-
angioplasty. Nephrol Dial Transplant 1998; 13(suppl 8):26–29. vasc Interv 2007; 69:1037–1043.
77. Bates MC, Campbell JE, Broce M, et al. Serum creatinine stabi- 97. Pillay WR, Kan YM, Crinnion JN, et al. Prospective multicentre
lization following renal artery stenting. Vasc Endovascular Surg study of the natural history of atherosclerotic renal artery steno-
2008; 42:40–46. sis in patients with peripheral vascular disease. Br J Surg 2002;
78. Thatipelli M, Misra S, Johnson CM, et al. Renal artery stent 89:737–740.
placement for restoration of renal function in hemodialysis 98. Mailloux LU, Bellucci AG, Mossey RT, et al. Predictors of
recipients with renal artery stenosis. J Vasc Interv Radiol 2008; survival in patients undergoing dialysis. Am J Med 1988; 84:
19:1563–1568. 855–862.
79. Sterling KA, Tehrani T, Rudnick MR. Clinical significance and 99. Parildar M, Parildar Z, Oran I, et al. Nitric oxide and oxidative
preventive strategies for contrast-induced nephropathy. Curr stress in atherosclerotic renovascular hypertension: effect of
Opin Nephrol Hypertens 2008; 17:616–623. endovascular treatment. J Vasc Interv Radiol 2003; 14:887–892.
80. Navaneethan SD, Singh S, Appasamy S, et al. Sodium bicarbon- 100. Higashi Y, Sasaki S, Nakagawa K, et al. Endothelial function and
ate therapy for prevention of contrast-induced nephropathy: a oxidative stress in renovascular hypertension. N Engl J Med
systematic review and meta-analysis. Am J Kidney Dis 2009; 2002; 346:1954–1962.
53:617–627. 101. Pizzolo F, Mansueto G, Minniti S, et al. Renovascular disease:
81. Solomon R. The role of osmolality in the incidence of contrast- effect of ACE gene deletion polymorphism and endovascular
induced nephropathy: a systematic review of angiographic revascularization. J Vasc Surg 2004; 39:140–147.
contrast media in high risk patients. Kidney Int 2005; 68: 102. Symonides B, Chodakowska J, Januszewicz A, et al. Effects of the
2256–2263. correction of renal artery stenosis on blood pressure, renal function
82. Liss P, Eklof H, Hellberg O, et al. Renal effects of CO2 and and left ventricular morphology. Blood Press 1999; 8:141–150.
iodinated contrast media in patients undergoing renovascular 103. Uzu T, Takeji M, Yamada N, et al. Prevalence and outcome of
intervention: a prospective, randomized study. J Vasc Interv renal artery stenosis in atherosclerotic patients with renal dys-
Radiol 2005; 16:57–65. function. Hypertens Res 2002; 25:537–542.
83. Zagler A, Azadpour M, Mercado C, et al. N-acetylcysteine and 104. Kennedy DJ, Colyer WR, Brewster PS, et al. Renal insufficiency
contrast-induced nephropathy: a meta-analysis of 13 randomized as a predictor of adverse events and mortality after renal artery
trials. Am Heart J 2006; 151:140–145. stent placement. Am J Kidney Dis 2003; 42:926–935.
84. Henry M, Klonaris C, Henry I, et al. Protected renal stenting with 105. Hunt JC, Sheps SG, Harrison EG Jr., et al. Renal and renovascular
the PercuSurge GuardWire device: a pilot study. J Endovasc Ther hypertension. A reasoned approach to diagnosis and manage-
2001; 8:227–237. ment. Arch Intern Med 1974; 133:988–999.
[2/7/010/20:55:16] [625–639]
106. Beek FJ, Kaatee R, Beutler JJ, et al. Complications during renal 125. AbuRahma AF, Stone PA, Bates MC, et al. Angioplasty/stenting
artery stent placement for atherosclerotic ostial stenosis. Cardio- of the superior mesenteric artery and celiac trunk: early and late
vasc Intervent Radiol 1997; 20:184–190. outcomes. J Endovasc Ther 2003; 10:1046–1053.
107. Hanzel G, Balon H, Wong O, et al. Prospective evaluation of 126. Boley SJ, Sprayregan S, Siegelman SS, et al. Initial results from an
aggressive medical therapy for atherosclerotic renal artery steno- agressive roentgenological and surgical approach to acute mes-
sis, with renal artery stenting reserved for previously injured enteric ischemia. Surgery 1977; 82:848–855.
heart, brain, or kidney. Am J Cardiol 2005; 96:1322–1327. 127. Gallego AM, Ramirez P, Rodriguez JM, et al. Role of urokinase in
108. Martin LG, Rundback JH, Sacks D, et al. Quality improvement the superior mesenteric artery embolism. Surgery 1996; 120:
guidelines for angiography, angioplasty, and stent placement in 111–113.
the diagnosis and treatment of renal artery stenosis in adults. 128. McBride KD, Gaines PA. Thrombolysis of a partially occluding
J Vasc Interv Radiol 2003; 14:S297–S310. superior mesenteric artery thromboembolus by infusion of strep-
109. Crea G, Rundback JH, Maddineni S. Managing complications of tokinase. Cardiovasc Intervent Radiol 1994; 17:164–166.
renovascular interventions. Tech Vasc Interv Radiol 1999; 2:98–106. 129. VanDeinse WH, Zawacki JK, Phillips D. Treatment of acute
110. Novick AC. Complications during renal artery stent placement mesenteric ischemia by percutaneous transluminal angioplasty.
for atherosclerotic ostial stenosis. J Urol 1998; 159:2245–2246. Gastroenterology 1986; 91:475–478.
111. National Heart Lung and Blood Institute. The Seventh Report of 130. Demirpolat G, Oran I, Tamsel S, et al. Acute mesenteric ischemia:
the Joint National Committee on Prevention, Detection, Evalua- endovascular therapy. Abdom Imaging 2007; 32:299–303.
tion, and Treatment of High Blood Pressure. Bethesda, MD: US 131. Lim RP, Dowling RJ, Mitchell PJ, et al. Endovascular treatment of
Department of Health and Human Services, National Institutes arterial mesenteric ischaemia: a retrospective review. Australas
of Health, 2004. Radiol 2005; 49:467–475.
112. Chi YW, White CJ, Thornton S, et al. Ultrasound velocity criteria 132. Lim RP, Dowling RJ, Thomson KR. Angioplasty and stenting of
for renal in-stent restenosis. J Vasc Surg 2009; 50:119–123. the superior mesenteric artery in acute mesenteric ischaemia.
113. Mohabbat W, Greenberg RK, Mastracci TM, et al. Revised duplex Australas Radiol 2004; 48:426–429.
criteria and outcomes for renal stents and stent grafts following 133. Nyman U, Ivancev K, Lindh M, et al. Endovascular treatment of
endovascular repair of juxtarenal and thoracoabdominal aneur- chronic mesenteric ischemia: report of five cases. Cardiovasc
ysms. J Vasc Surg 2009; 49:827–837; discussion 37. Intervent Radiol 1998; 21:305–313.
114. Raza SA, Chughtai AR, Wahba M, et al. Multislice CT angiog- 134. Landis MS, Rajan DK, Simons ME, et al. Percutaneous manage-
raphy in renal artery stent evaluation: prospective comparison ment of chronic mesenteric ischemia: outcomes after interven-
with intra-arterial digital subtraction angiography. Cardiovasc tion. J Vasc Interv Radiol 2005; 16:1319–1325.
Intervent Radiol 2004; 27:9–15. 135. Matsumoto AH, Angle JF, Spinosa DJ, et al. Percutaneous trans-
115. Rheudasil JM, Stewart MT, Schellack JV, et al. Surgical treatment luminal angioplasty and stenting in the treatment of chronic
of chronic mesenteric arterial insufficiency. J Vasc Surg 1988; mesenteric ischemia: results and longterm followup. J Am Coll
8:495–500. Surg 2002; 194:S22–S31.
116. Geelkerken RH, van Bockel JH, de Roos WK, et al. Chronic 136. Gentile AT, Moneta GL, Taylor LM Jr., et al. Isolated bypass to
mesenteric vascular syndrome. Results of reconstructive surgery. the superior mesenteric artery for intestinal ischemia. Arch Surg
Arch Surg 1991; 126:1101–1106. 1994; 129:926–931; discussion 31–32.
117. Harward TR, Brooks DL, Flynn TC, et al. Multiple organ dys- 137. McAfee MK, Cherry KJ Jr., Naessens JM, et al. Influence of
function after mesenteric artery revascularization. J Vasc Surg complete revascularization on chronic mesenteric ischemia. Am
1993; 18:459–467; discussion 67–69. J Surg 1992; 164:220–224.
118. Moawad J, Gewertz BL. Chronic mesenteric ischemia. Clinical 138. Foley MI, Moneta GL, Abou-Zamzam AM Jr., et al. Revascula-
presentation and diagnosis. Surg Clin North Am 1997; 77:357–369. rization of the superior mesenteric artery alone for treatment of
119. Connolly JE, Kwaan JH. Management of chronic visceral ische- intestinal ischemia. J Vasc Surg 2000; 32:37–47.
mia. Surg Clin North Am 1982; 62:345–356. 139. Johnston KW, Lindsay TF, Walker PM, et al. Mesenteric arterial
120. Dick AP, Graff R, Gregg DM, et al. An arteriographic study of bypass grafts: early and late results and suggested surgical
mesenteric arterial disease. I. Large vessel changes. Gut 1967; approach for chronic and acute mesenteric ischemia. Surgery
8:206–220. 1995; 118:1–7.
121. Morris GC Jr., De Bakey ME, Bernhard V. Abdominal angina. 140. Mateo RB, O’Hara PJ, Hertzer NR, et al. Elective surgical
Surg Clin North Am 1966; 46:919–930. treatment of symptomatic chronic mesenteric occlusive disease:
122. Cunningham CG, Reilly LM, Stoney R. Chronic visceral ische- early results and late outcomes. J Vasc Surg 1999; 29:821–831;
mia. Surg Clin North Am 1992; 72:231–244. discussion 32.
123. Rose SC, Quigley TM, Raker EJ. Revascularization for chronic 141. Kougias P, El Sayed HF, Zhou W, et al. Management of chronic
mesenteric ischemia: comparison of operative arterial bypass mesenteric ischemia. The role of endovascular therapy. J Endo-
grafting and percutaneous transluminal angioplasty. J Vasc vasc Ther 2007; 14:395–405.
Interv Radiol 1995; 6:339–349. 142. Reuter SR. Accentuation of celiac compression by the median
124. Kasirajan K, O’Hara PJ, Gray BH, et al. Chronic mesenteric arcuate ligament of the diaphragm during deep expiration. Radi-
ischemia: open surgery versus percutaneous angioplasty and ology 1971; 98:561–564.
stenting. J Vasc Surg 2001; 33:63–71.
[2/7/010/10:21:11] [640–658]
51
Carotid and vertebral artery interventions

Marco Roffi
INTRODUCTION the apex of the aortic arch so that a horizontal line drawn
In Western countries, stroke is the third commonest cause of perpendicular to the long axis of the human body at the apex of
death, behind cardiac disease and cancer, and is the number the arch will intersect the origin of all three vessels (type I arch)
one condition associated with serious, long-term disability (1). (Fig. 51.2). However, elongation and rostral migration of the
The yearly incidence of stroke approaches 0.2% of the Western distal aortic arch with increasing age, atherosclerotic burden,
population and the number of stroke-related deaths is expected and hypertension may lead to a change in the relative positions
to double over the next 30 years. In the United States, each year of the great vessels: the left common carotid artery (CCA) and
approximately 800,000 people experience a new or recurrent the left subclavian artery (SCA) migrate rostrally along with the
stroke and for the year 2009, the estimated direct and indirect distal arch, which takes on a narrowed and peaked appearance
costs of stroke are estimated at $69 billion (1). A stenosis of the rather than a smooth convex shape. This leads to the brachioce-
internal carotid artery (ICA) is the underlying condition in 10% phalic trunk (or innominate artery) arising “lower” and appear-
to 20% of patients presenting with ischemic stroke (2). ing to arise from the ascending aorta, followed by the left CCA
Large-scale randomized clinical trials have established and then the left SCA. This constellation makes cannulation of
the superiority of carotid endarterectomy (CEA) over medical the brachiocephalic trunk and left CCA difficult if not impossible
management in patients with high-grade carotid stenosis, par- (so-called type III arch) (Fig. 51.2). The cannulation of the left
ticularly in the presence of symptoms. However, these results CCA may also be challenging in the presence of the so-called
were obtained by high-volume surgeons on selected patients “bovine arch.” In this configuration, only the brachiocephalic
and may therefore not be reproducible in clinical practice. trunk and the left SCA arteries arise from the arch, while the left
Although balloon angioplasty of carotid stenosis was per- CCA originates from the brachiocephalic trunk (Fig. 51.2).
formed since the early 1980s, the use of stents in the mid- Once the CCA is cannulated, it is important to clearly
1990s and of emboli protection devices (EPDs) around the year differentiate the ICA from the external carotid artery (ECA). The
2000 have led to an exponential growth of carotid artery most important difference between the two is that the ICA has
stenting (CAS) procedures performed worldwide. In the United no branches in the extracranial portion, while the ECA—smaller
States, a nationwide survey including over 130,000 Medicare in caliber—has extensive early branching. Subsequently, baseline
patients undergoing carotid artery revascularization showed intracranial imaging of the carotid artery needs to be performed
that the overall volumes of procedures decreased by 11% to assess the presence of collaterals, as described in chapter 26,
between 1998 and 2004 (3). In this period of time, while the and have a baseline angiographic status.
number of CEA decreased by 17%, the number of CAS
increased by 149% (Fig. 51.1). However, and possibly related
to limited Medicare reimbursement for CAS, the overall pro- Cerebral Angiography
portion of patients treated by endovascular approach remained The fundamentals of cerebral angiography are reported in
low (3.8% in 1998 and 10.5% in 2004). chapter 26. In the absence of contraindications—such as renal
A stenosis of the vertebral artery (VA) rarely causes insufficiency, heart failure, or high-risk aortic arch anatomy—
symptoms and, as a consequence, revascularization is seldom we encourage the performance of a four-vessel cerebral angiog-
required. While surgery is in most centers not considered a raphy—defined as the selective engagement of both CCA and
viable option for VA stenosis, the number of endovascular the nonselective engagement of at least one VA prior to starting
vertebral procedures remain limited compared to CAS proce- the CAS procedure. The purpose of a complete angiographic
dures. Although the main focus of the chapter is on carotid work-up is the assessment of the entire extra- and intracranial
artery interventions, the specifics of vertebral interventions will cerebrovascular vasculature to identify the collateral circulation
be mentioned. and to exclude associated vascular conditions such as aneur-
ysms or arteriovenous malformations. Finally, routine perfor-
mance of cerebral angiography enhances the catheterization
ANATOMIC CONSIDERATIONS skills and the understanding of neurovascular anatomy.
Aortic Arch and Carotid Arteries The complication rates of diagnostic angiography have
The details of carotid and cerebral angiography are described significantly decreased over the last two decades, likely due to
in detail in chapter 26. The present section recalls just the improvements in equipment and operators’ skills. Important
anatomic notions critical for the planning and the performance measures to increase safety include periprocedural anticoagula-
of CAS. The vertebral circulation, not covered elsewhere in the tion and monitoring of the catheter-tip pressure throughout the
book, is addressed in more detail. Acquired abnormalities of procedure in order to prevent plaque dislodgement during cath-
the aortic arch are common and may increase the difficulty of eter manipulation or dye injection. A recent prospective series of
carotid cannulation. Normally, all three great vessels arise from almost 3000 diagnostic cerebral angiographies astonishing low
[2/7/010/10:21:11] [640–658]
CAROTID AND VERTEBRAL ARTERY INTERVENTIONS 641
Figure 51.1 Total number of carotid revascularization

procedures performed in the United States on Medicare
beneficiaries (MCBEs). Source: From Ref. 3.
Figure 51.2 Aortic arch configurations as demonstrated on aortic arch angiography in 408 left anterior oblique projection. On the left panel,
favorable angles to be overcome in order to cannulate the common carotid arteries (CCA) in a type I arch. On the middle panel, a type III arch
configuration makes the cannulation of the CCA more challenging. On the right panel, a “bovine” arch configuration makes the cannulation of
the left CCA more difficult.
rates of neurologic complications with 0.3% transient ischemic and 25% of cases, respectively. In the remaining cases, the two
attacks (TIA) and no permanent deficit (4). Similarly, an analysis vessels have similar caliber. In approximately 15% of the pop-
of over 19,000 consecutive patients detected a permanent neuro- ulation one VA is atretic—that is, it has a diameter <2 mm—
logic deficit rate following cerebral angiography of 0.14% (5). and supplies only the posterior inferior cerebellar artery (PICA)
or gives only minimal contribution to the basilar system. The
VA is divided into four segments (Fig. 51.3):
The Vertebral Arteries
The VA is usually the first branch off the SCA artery. On rare l The V1 segment extends from the VA origin to the trans-
occasions, the left VA arises directly from the aortic arch, verse foramen of the sixth cervical (C6) vertebra; it has no
between the left CCA and left SCA or distally to the left SCA. branches.
The diameter of the VA is 3 to 5 mm, and in the same patient, l The V2 segment extends from the transverse foramen of C6
the caliber may vary to a great extent between the left and the to C1; it gives off the meningeal, muscular, and radicular
right side. The left and the right VA are dominant in over 50% arteries.
[2/7/010/10:21:11] [640–658]
placed in the SCA is usually sufficient for diagnostic purposes.

In order to improve the quality of the subselective injection, we
recommend the inflation of a blood pressure cuff at the ipsi-
lateral arm at suprasystolic pressures.
The left SCA and the brachiocephalic trunk are usually
easily accessible with a variety of diagnostic catheters, includ-
ing Judkins right 4, vertebral/Bernstein or headhunter. In more
complex arches, the brachiocephalic trunk may be cannulated
using catheter shapes such as Benson, Simmons/Sidewinder,
or Vitek. A roadmap is than performed, and the catheter is
advanced over a steerable 0.035-in. wire in the SCA.
Selective VA angiography requires gentle manipulation
and for that purpose simple-shaped diagnostic catheters—such
as the vertebral, Judkins right or headhunter—are preferred.
Frequently, positioning the catheter at the ostium may allow
selective angiography. If this is not the case and selective
angiography is required, a steerable floppy 0.035-in. guidewire
is positioned at the mid-distal third of the V2 segment avoiding
hooking in small branches and then the catheter is advanced in
the VA over the wire. For diagnostic purposes, the catheter
should be kept in the V1 segment. In difficult anatomical
situations, the selective catheterization of the VA should be
Figure 51.3 Digital subtraction angiography in anteroposterior discouraged because of the associated dissection risk. Vessel
view showing the four segments of the left vertebral artery (V1 to wall injury may be the result of catheter manipulations or
V2 in the left panel and V2 to V4 in the right panel). Abbreviation: forced contrast injection.
SCA, subclavian artery.
INDICATIONS
Carotid Artery Stenting
The presence of a carotid bruit is neither sensitive nor specific
for the presence of a significant stenosis of the ICA. The
l The V3 segment extends from the transverse foramen of imaging modality of choice in patients with suspected carotid
C1 to the foramen magnum, where it penetrates the dura; stenosis is duplex ultrasound. If a significant stenosis of the
it provides small meningeal branches. ICA is detected, the finding should be confirmed either by CT
l The V4 segment extends from the entrance of the VA angiography or MR angiography. At the same time, the brain
through the dura to the level of medullopontine junction, should be imaged for the detection of silent ischemic lesions
where it joins the other VA to form the basilar artery; the and to rule out associated intracranial pathologies such as brain
major branches are the anterior and posterior spinal tumors, arteriovenous malformations, or vascular aneurysms.
arteries as well as the largest branch of the VA, the CAS is an appealing alternative to CEA because it is less
PICA. In some instances, the VA may terminate at the invasive and it can potentially address lesions that may not
PICA and not contribute to basilar artery flow. be treated surgically. With respect to the degree of stenosis to
The circulation at the level of the circle of Willis is detailed in be treated in asymptomatic and symptomatic patients, CAS
chapter 26. With respect to the posterior circulation, the most and CEA share the same indications. The widely accepted
common variants are hypoplastic or absent posterior commu- guidelines of the American Heart Association support carotid
nicating artery (PCom), unilaterally or bilaterally. The other revascularization—and specifically CEA—for patients with
common variant, the so-called fetal posterior cerebral artery symptomatic carotid stenosis 50% and an estimated perioper-
(PCA) defines the persistence of a large PCom that feeds the ative death or stroke risk <6% and for asymptomatic stenosis
entire PCA. In this variant, the PCA is a branch of the ICA. 60%, as long as the estimated perioperative death or stroke is
<3% and life expectancy is 5 years (6). However, due to the
usually benign course of asymptomatic carotid disease, in
clinical practice, asymptomatic stenoses are frequently treated
FUNDAMENTALS only beyond a stenosis grade of 80%.
The technique of diagnostic carotid angiography, usually per- Once the indication for revascularization is established,
formed in local anesthesia and using a femoral approach, is further steps in decision making should be based on the sur-
detailed in chapter 26. The administration of an anticoagulant— gical and endovascular risk of the patient and on the locally
commonly unfractionated heparin—prior to cannulation of the available CAS expertise (Fig. 51.4). Importantly, the conditions
supra-aortic vessels is recommended, particularly if a complete associated with poor outcomes for CEA and CAS differ.
cerebral angiogram is planned. With respect to VA angiography, Accordingly, while the outcomes of CEA are greatly influenced
in order to delineate the intracranial circulation, the injection of by the comorbidities of the patient, poor outcomes with CAS
one VA—the dominant one—is sufficient. Since cannulation of are related to challenging anatomies at the level of the aortic
the VA may be associated with higher complication rates than arch and of the carotid arteries. While advantages and disad-
CCA cannulation, selective VA angiography is not routinely vantages of CEA and CAS are listed in Table 51.1, the (relative)
performed. Subselective VA opacification with the catheter contraindications to CAS are reported in Table 51.2. The
[2/7/010/10:21:11] [640–658]
Figure 51.4 Management algorithm for patients requiring carotid revascularization. Abbreviations: CEA, carotid endarterectomy; CAS,
carotid artery stenting; EPD, embolic protection devices. Source: From Ref. 7.
Table 51.1 Pros and Cons of Carotid Revascularization Procedures

Endarterectomy Stenting
Pros Widely available Outcome less influenced by comorbidities
Excellent results for high-volume surgeons/ Local anesthesia
hospitals in low-risk patients
No neck incision/scar
Usually next day discharge
Cons Outcome influenced by comorbidities Fewer experienced operators
Frequently performed in general anesthesia Risk of the procedure may increase in patients with
Neck incision/scar - severe peripheral vascular disease,
Neck complications, cranial nerve palsies - severely calcified, tortuous/steep aortic arch, and
Not suitable for high or low carotid lesions - severe calcification or tortuosity of cervicocranial vessels
Longer hospital stay Femoral access site complications
May not be performed if aspirin/clopidogrel intolerance
greatest advantage of CAS over CEA appears to be in patient setting for CAS is the steep aortic arch with or without severe
with restenosis post-CEA, following neck radiation or neck tortuosity of the common or ICA (Fig. 51.2). As a general rule, if
dissection, contralateral carotid occlusion, or surgically poorly unexpected difficulties are encountered during engagement of
accessible carotid lesions. CAS may also be considered for the CCA because of (underestimated) anatomic challenges, it is
patients who recently underwent coronary drug-eluting stent preferable to abort the procedure and to consider surgery. The
implantation, as any surgical procedure including CEA—even inability to use an EPD, a rare occurrence based on the different
if performed under dual antiplatelet therapy with aspirin and modalities available, should also be considered a relative con-
clopidogrel—may be associated with an increased risk of cor- traindication to CAS. Additional relative contraindications to
onary stent thrombosis. CAS include severe circumferential calcification of the carotid
Surgery should be preferred in the presence of severe bulb, which may limit the ability to dilate the lesion and the
peripheral arterial disease not allowing for femoral arterial presence of a long (>15 mm), tubular lesions in the carotid
access. Although CAS can be performed via the brachial or bulb, which may increase the risk of distal embolization. More-
radial approach, the procedure is more demanding and likely over, patients with renal insufficiency or a history of allergic
associated with higher complication rates. Another challenging reactions to angiographic contrast may not be appropriate
[2/7/010/10:21:11] [640–658]
Table 51.2 Conditions Associated With Increased Risk Procedural some series in terms of periprocedural stroke, surgery has been
Risk and Contraindications for Carotid Artery Stenting associated with a high frequency of non-stroke-related compli-
cations, such as Horner’s syndrome, lymphatic injury, VA
Increased risk
thrombosis, and laryngeal nerve injury (11).
Age 80 years
Symptomatic ICA lesion The most recent American Heart Association stroke pre-
Severe renal insufficiency vention guidelines state that endovascular treatment of patients
Severely diseased and/or steep aortic arch with symptomatic extracranial vertebral stenosis may be con-
Severely diseased and/or tortuous CCA sidered when patients remain symptomatic despite optimal
Severely diseased and/or tortuous distal ICA medical management (Class IIb, Level of Evidence C) (12). As
Long subtotal ICA occlusion (string sign) with any interventional procedure, the risk of the intervention
Poor femoral access must be outweighed by the potential benefits of revasculariza-
Major stroke within 4–6 wk tion. The need to intervene is tempered by the fact that the
Extensive intracranial microvascular disease
posterior circulation is supplied by the confluence of the two
Contraindications VA and a large proportion of patients remain asymptomatic
Intolerance to aspirin and/or clopidogrel despite occlusion of one extracranial VA.
Circumferential ICA calcification
Patients with isolated stenosis of the extracranial VA >
Intraluminal thrombus
50% confirmed on angiography and symptoms attributed to
Chronic ICA occlusion
Intracranial aneurysm or AVM requiring treatment vertebrobasilar ischemia may benefit from stenting, especially
if the vessel is dominant or the contralateral site occluded. The
Abbreviations: CCA, common carotid artery; ICA, internal carotid majority of patients with asymptomatic extracranial VA disease
artery; AVM, arteriovenous malformation.
do not require treatment. Asymptomatic patients with high-
grade (>70% stenosis) lesions may benefit from revasculariza-
tion, especially if the severity of the lesion is progressive and
the affected VA is dominant or is the only patent one.
candidates for CAS. Finally, for patients who cannot tolerate
dual antiplatelet therapy with aspirin and clopidogrel (e.g., in
EQUIPMENT
the presence of bleeding diathesis, intracerebral hemorrhage or
In addition to an appropriate angiography suite with digital
cerebral arteriovenous malformation, documented intolerance
subtraction and roadmap capabilities, the equipment required
to one of the agents, or noncompliance) surgery may be
for CAS includes diagnostic catheters, sheath systems, guide
preferred.
catheters, guidewires, EPDs, balloons catheters, and stents.
The locally available surgical and endovascular expertise
Since for every device manipulation there is a learning curve,
are also important parameters in decision making. The term
it is recommended at the beginning of the CAS experience to be
“dedicated CAS centers” defines hospitals with multidiscipli-
familiar with a limited basic armamentarium, which can be
nary programs for the management of patients with carotid
subsequently expanded (Table 51.3). The basic equipment
artery stenosis, documented CAS expertise with a track record
should include few diagnostic catheters, a 8-Fr guiding catheter
of complications within the guideline’s limits, and imple-
or 90-cm long sheath, three 0.035-in. wires (steerable, hydro-
mented prospective quality control programs. Hospitals not
philic stiff, and stiff in exchange length), one balloon catheter
fulfilling these criteria should treat carotid patients with CEA
type, one nitinol stent type, and one or two filter-based EPD
and in case of high or prohibitive surgical risk, transfer them to
systems. If a 8-Fr guiding catheter is used, we encourage
a referral center.
advancing it over a 5-Fr, 125-cm long diagnostic catheter with
Reimbursement issues do influence the practice of CAS.
Judkins right or multipurpose shapes (telescoping technique).
While in Europe, CAS is reimbursed in most but not all
Rarely, the filter EPD may get clogged during the procedure as
countries, in the United States the Centers for Medicare and
a consequence of distal embolization, with subsequent no flow
Medicaid Services (CMS) have concluded that CAS with EPD is
reasonable and appropriate for symptomatic patients with
carotid stenosis 70% at high risk for surgery (10). If per-
formed within clinical trials or CAS postapproval studies, the
Table 51.3 Minimal Equipment to Start Carotid Artery Stenting
procedure may be performed in high-risk symptomatic patients
l Angiographic suite with digital subtraction angiography (DSA)
with 50% to 70% stenosis as well as in high-risk asymptomatic
individuals with stenosis 80%. Carotid stenting is considered and roadmap capabilities
l Diagnostic catheters (5 Fr)
appropriate if performed in facilities and by operators comply-
l JR4 or vertebral, Headhunter, Vitek, or Simmons/
ing with the set standards (10).
Sidewinder 1, 2
l 0.035-in. guidewires
Vertebral Interventions l Steerable, hydrophilic stiff, and stiff (exchange length)
l 8-Fr guiding catheter (headhunter or multipurpose shape)
Treatment of atherosclerotic extracranial VA occlusive disease
or 6-Fr, 90-cm sheath
can consist of medical, surgical, or endovascular therapies. l Filter emboli protection device
While the optimal antithrombotic therapy—aspirin, clopidog- l 0.014-in. balloon catheter, rapid exchange, and approved
rel, or warfarin—in VA stenosis has not been defined, most for CAS
patients are treated with low-dose aspirin. Surgery—including l 0.014-in. nitinol self-expanding stent, and rapid exchange
vertebral endarterectomy and bypass operations—is technically l Coronary aspiration catheter
very challenging and is not considered a viable option in most l Femoral closure devices
centers. Accordingly, despite having favorable success rates in
[2/7/010/10:21:11] [640–658]
in the ICA. In these circumstances, a coronary aspiration cath- Once cannulation of the CCA is completed, the CAS
eter such as Export (Medtronic, Santa Rosa, California, U.S.) or procedure is continued over a 0.014-in. guidewire, usually
Diver (Invatec, Roncadelle, Italy) should be used to aspirate the incorporated in the EPD. For cases performed with proximal
blood column in the internal carotid prior to EPD removal. occlusion EPD or with filter systems that allow use of separate
With increasing CAS experience, the goal should be to tailor the preferred guidewires, a variety of 0.014-in. coronary guide-
equipment to the patient and the lesion. To prevent femoral wires may be used, such as the Balanced Middle Weight or
access bleeding in those patients aggressively anticoagulated Balanced Heavy Weight (Abbott) or the Stabilizer (Cordis). The
we encourage the use of closure devices. same wires can also be used as additional “buddy wires” in the
presence of excessive tortuosity of the ICA. Unprotected CAS
(not recommended) may also be performed over 0.018 in.
Sheath Systems peripheral guidewires such as Steelcore (Abbott), V18 Control
Several sheath systems, usually 90-cm long, are available to
(Boston Scientific), Roadrunner (Cook), and SV (Cordis).
access the CCA. Important features include kinking resistance
and flexibility, a radiopaque band at the tip of the sheath to
allow for accurate positioning, an atraumatic soft tip to reduce Emboli Protection Devices
vessel trauma during engagement, and hydrophilic coating for Three types of EPDs are currently on the market: filter-based,
enhanced trackability. Frequently used devices include the Cook distal balloon occlusion, and the proximal occlusion (flow
Shuttle sheath (Cook Inc., Bloomington, Indiana, U.S.), the reversal) EPD (Fig. 51.5). Additional details on EPDs are pro-
Avanti and the Brite Tip sheath systems (Cordis Corporation, vided in chapter 37. A large prospective registry reported that
Miami Lakes, Florida, U.S.), and the Pinnacle Destination guid- filter-based EPD is the currently preferred method of emboli
ing sheath (Terumo Medical Corporation, Somerset, New Jersey, protection for CAS in clinical practice and both filter-based and
U.S.). We recommend the use of 6-Fr sheath despite the fact that distal occlusion EPD seem to be equally effective (13). Advan-
some of the stent systems may be 5-Fr sheath compatible. tages and disadvantages of the different device types are
Accordingly, a tight fit of the stent system in the sheath bears
the risk of air embolization at the time of device advancement.
Guide Catheters
Guide catheters may be used to cannulate the CCA as alterna-
tive to long sheaths. The most frequently used shapes are
the headhunter (H1, Cordis or Cook) and the multipurpose.
For the same reasons as described above, we recommend 8-Fr
guide catheters although some stent systems may be delivered
through 7 Fr guides. The choice of performing CAS using a
long sheath or a guide catheter is matter of personal preference
and training. Overall, the guide catheter provides better torque
control and stability and lesser chance of kinking but requires
larger access size. In addition, during advancement of the guide
catheter over a 0.035-in. guidewire, the abrupt transition at the
tip may predispose to scraping the vessel wall with subsequent
distal embolization. Therefore, we recommend to always
advancing an 8-Fr guide over a 125-cm long 5-Fr diagnostic
catheter (coaxial or telescoping technique). As a general rule,
the more complex the anatomy at the level of the aortic arch
and origin of the supra-aortic vessels, the more advantageous
may be the use of a guide catheter over a sheath. Figure 51.5 Strategies for emboli protection in carotid artery
stenting. On the left panel is demonstrated a filter device, in the
middle a distal balloon occlusive device, and in the right panel a
Guidewires proximal occlusive device. Filter devices incorporate an angioplasty
Several 0.035-in. guidewires are available for selective CCA guidewire with a filter that expands and is placed distal to the lesion
cannulation for diagnostic and interventional purposes, includ- to capture and retrieve embolic debris that should get detached
ing steerable guidewires with soft tip such as the Wholey during the intervention. At the end of the procedure, the filter is
(Mallinckrodt, St. Louis, Missouri, U.S.), Magic Torque (Boston collapsed and removed from the artery. Distal balloon occlusion
devices (middle panel) result in complete occlusion of antegrade
Scientific, Natick, Massachusetts, U.S.), Storq (Cordis) wires, or
flow in the internal carotid artery (ICA) while balloon angioplasty and
hydrophilic wires such as the Glidewire (Terumo, Westwood, stenting are performed. The entire blood column—possibly contain-
Massachusetts, U.S.). For excessively tortuous arteries and ing embolic debris—is then aspirated before restoration of the
challenging aortic arches, a stiff hydrophilic wire (e.g., stiff circulation. The principle of proximal occlusion devices (right
Glidewire, Terumo) may be used to advance the diagnostic panel) is based on simultaneous occlusion of the external carotid
catheter. If the initial wire needs to be exchanged (over the artery (ECA) and of the common carotid artery (CCA). Depending
diagnostic catheter) for an extrastiff one, then the options on the devices used, during the stenting procedure, the flow in the
include the Supracore (Abbott Vascular, Redwood City, Cali- ICA is either reversed (Gore Neuro Protection; Gore Medical, Flag-
fornia, U.S.) and the Amplatz guidewires (Boston Scientific or staff, Arkansas, U.S.) or arrested and then aspirated (MO.MA,
Cook) in exchange length. Typically the stiff guidewire is Invatec, Roncadelle, Italy).
placed in the ECA over a 5-Fr diagnostic catheter.
[2/7/010/10:21:11] [640–658]
Table 51.4 Pros and Cons of Emboli Protection Devices

Device type Pros Cons
Filters Preserve antegrade flow throughout the procedure May not capture debris smaller than pore size
Optimal visualization of the lesion Not as steerable as coronary guidewiresa
Lesion crossing with guidewire of choice possible May cause spasm or dissection of the internal
(with wire-independent systems) carotid artery
Can be deployed and captured rapidly Lesion crossing not protected
Easy to use May cause flow obstruction
May not be placed in the presence of excessive
ICA tortuosity
Apposition in tortuous vessel may be suboptimal
Proximal balloon occlusion All the steps of the procedures protected Transient flow obstruction may be poorly
flow reversal toleratedb
Crossing of the lesion with guidewire of choice Poor visualization of the lesion
Protection possible also in the presence of Handling more demanding than filter EPD
excessive tortuosity of the internal carotid artery
Protection independent of particle size Larger sheath size required
Occlusive balloons may cause dissection or
spasm of the common or external
carotid artery
Time-consuming setup
Distal balloon occlusion Protection independent of particle size Transient flow obstruction may be poorly
toleratedb
Lower profile and less stiff than filter EPD Poor visualization of the lesion
More easily delivered in tortuous anatomy Crossing of the lesion not protected
Use more cumbersome than filter EPD
Potential for balloon-induced injury
Less steerable than coronary guidewires
a
Does not apply for wire-independent systems.
b
Problematic in patients with severe stenosis or occlusion of the contralateral ICA or isolated hemisphere.
Abbreviations: EPD, emboli protection devices; ICA, internal carotid artery.
reported in Table 51.4. Filter-based EPD usually incorporate a

0.014-in. guidewire with a filter that expands and is placed
distal to the target lesion to capture and retrieve embolic debris
that may get detached during balloon angioplasty and stenting.
The filtering is performed by a polyurethane membrane with
laser-drilled holes or a nitinol mesh (Fig. 51.6). At the end of the
procedure, the filter is collapsed, trapping the embolic debris,
and is removed from the artery. The major advantage of filter
EPD is that the perfusion is maintained throughout the proce-
dure. A variety of filter-based EPD is currently on the market,
and the technical details are reported in Table 51.5. In the
majority of the devices, the filter EPD is mounted directly on
a wire (wire dependent). Few of them, such as the Emboshield
Pro (Abbott) or Spider (ev3), allow for the advancement of a
wire followed by the device (wire independent). This may be of
advantage for complex lesions. In contrast to filter EPD, the use
of balloon occlusive devices results in complete occlusion of
antegrade flow while the device is deployed (Table 51.4). With
distal balloon occlusion EPD (PercuSurge GuardWire, Med-
tronic), the wire can be used as an angioplasty guidewire and
provides protection from distal embolization by means of an
occlusion balloon. An inflation device allows controlled expan-
sion of the balloon in the treated vessel. An aspiration catheter
is used to remove the debris from the treated vessel before the
balloon is deflated and antegrade flow in the treated vessel is Figure 51.6 Two examples of filter emboli protection devices. In
restored. Another approach for EPD is to occlude the inflow to the upper and lower panels are demonstrated the Spider (ev3,
the brain by using a proximal occlusion balloon in the CCA. Plymouth, Minnesota, U.S.) and the Emboshield Pro (Abbott Vas-
The proximal balloon occlusion systems include the MO.MA cular, Redwood City, California, U.S.), respectively.
(Invatec) and the Gore Neuro Protection (Gore Medical, Flagstaff,
[2/7/010/10:21:11] [640–658]
Table 51.5 Comparison of Selected Distal Emboli Protection Filters

Characteristic Spider Filterwire Angioguard Accunet Emboshield Pro Interceptor
Manufacturer ev3 BSC CJJ GDT ABT MDT
Material N N, PU N, PU N, PU N, PU N
Guidewire (in.) 0.014 0.014 0.014 0.014 0.014 0.014
RX Yes Yes Yes Yes Yes Yes
Independent wire Yes No No No Yes No
Sheath (Fr) 6 6 6 6 6 6
Vessel size (mm) 3.0–7.0 3.5–5.5 4.0–8.0 4.5–7.5 4.0–7.0 4.5–6.5
Profile (Fr) 3.2 3.2 3.2–3.9 3.5–3.7 3.2 2.7
Pore size (m) 167–209 110 100 150 140 100
FDA approval Yes Yes Yes Yes Yes Yes
Abbreviations: ABT, Abbott Medical Corporation; BSC, Boston Scientific Corporation; CJJ, Cordis/Johnson & Johnson Inc.; FEP, fluorinated
ethylene propylene; Fr, French; GDT, Guidant Corporation; GFS, Gore Filter System; MDT, Medtronic Corporation; N, nitinol; PTFE,
polytetrafluoroethylene; PU, polyurethane; RX, rapid exchange; FDA, United States Food and Drug Administration.
Arkansas, U.S.) and are based on a guiding catheter with an Although differences in stent designs are interesting from a
occlusion balloon attached at its distal end that is inflated in the conceptual standpoint, their clinical impact remains to be
CCA. Protection is established by flow reversal (Gore Neuro demonstrated. Therefore, at the beginning of the learning
Protection) or flow arrest (MO.MA) in the ICA. At the same curve, we recommend to be familiar with one device instead
time, the ECA is occluded to avoid retrograde collateral flow of tailoring the stent for a specific patient or lesion.
from the ECA into the ICA. With the MO.MA device, the
particles are removed by aspiration of the blood column fol-
lowing stenting and before ECA and CCA balloon deflation.
Equipment for Vertebral Interventions
Most vertebral interventions can be performed with coronary
equipment using a 6-Fr guiding catheter, a medium or extra
Balloon Catheters support 0.014-in. coronary guidewire, and a balloon-expandable
As a general rule, the balloon catheter used for predilatation coronary or peripheral stent (16). Guiding catheter shapes
(optional) should be greatly undersized and the one chosen for
postdilatation (mandatory) should be shorter than the stent and
also slightly undersized. To minimize the risk of distal emboli-
zation, a residual stenosis of up to 30% following postdilatation
of the stent is acceptable. Typically, balloon catheters 3.0 to 4.0
mm in diameter and 20 mm in length are used for predilatation,
while the balloon size for postdilatation of the stent is usually 5.0
to 5.5 mm in diameter and 20 mm in length. The use of 0.014-in.
rapid exchange balloon catheters is recommended. For medical-
legal reasons, it is preferable to use balloon catheters, which are
labeled for carotid interventions, including Avion (Invatec),
Aviator (Cordis), Ultra-Soft SV, and Sterling (Boston Scientific).
Stents
In order to prevent stent crushing, the devices used in the
carotid territory are all self-expanding (Fig. 51.7). With the
exception of the stainless steel Carotid Wallstent (Boston Sci-
entific), all the devices are based on nitinol, an alloy allowing
for excellent flexibility, conformability, and crush resistance.
The properties of some of the carotid stents on the market are
listed in Table 51.6. With respect to stent design, the most
important differentiation is between open- and closed-cell
designs. Closed-cell stents are characterized by fully connecting
struts and may offer greater plaque coverage than open-cell
stents—with both connecting and nonconnecting struts—
potentially reducing the risk of delayed embolism due to
plaque protrusion. Therefore, this type of stents may be pre-
ferred for soft or ulcerated plaques with high embolic potential.
Finally, closed-cell stents have higher radial strength, an impor-
tant feature in severely calcified lesions. Open-cell stents are Figure 51.7 Self-expanding nitinol carotid stent (Cordis Corpora-
more flexible, conform better to the artery, and result in better tion, Miami Lakes, Florida, U.S.). The lower panel demonstrates the
wall apposition. Therefore, this stent type may be preferred in flexibility conveyed by this alloy.
tortuous arteries and in lesions extending into the CCA.
[2/7/010/10:21:11] [640–658]
Table 51.6 Design and Brand Names of Commonly Used Stents increased risk of embolization. In the presence of a bovine arch,
direct intubation of the left CCA may be achieved with a coro-
Stent type Name Company FDA approval
nary Amplatz left catheter or a Simmons/Sidewinder or Benson
Open-cell Acculink Abbott 3 catheter. The details of diagnostic angiography techniques are
nitinol discussed in chapter 26. The intubation techniques with different
Exponent Medtronic 3 catheters can be successfully learnt at simulators (18).
Precise Cordis 3
Protege ev3 3
Vivexx Bard 8 Carotid Artery Stenting
Closed-cell NexStent Boston 3 On the basis of diagnostic angiography, the strategy and equip-
nitinol Scientific ment should be determined and prepared to minimize the
Xact Abbott 3 procedural time. Once diagnostic angiography is performed
Hybrida Cristallo Invatec 8
and it is assessed that the anatomy is suitable for CAS, addi-
Stainless Wallstent Boston 3
steel Scientific tional unfractionated heparin should be administered, if
a
needed, to achieve an activated clotting time of 250 to 300
Hybrid design since in part open cell in part closed cell. seconds. Alternatively, bivalirudin has been used as anticoa-
Abbreviation: FDA, United States Food and Drug Administration.
gulant for CAS. Patients not on chronic aspirin require 250 to
500 mg of aspirin administered orally or intravenously prior to
the procedure. For patients not on chronic clopidogrel therapy,
frequently used include the Judkins right or vertebral shapes.
we recommend 300 mg of clopidogrel the day before the
Since the lesions are frequently in ostial location, coronary stents
procedure. Documented aspirin or clopidogrel intolerance
with good radial strength are preferred. EPDs are usually not
should be considered a contraindication for CAS, and patients
recommended. In fact, although embolization may occur during
should be referred for surgery. To allow for assessment of
vertebral interventions, the use of EPD in the vertebral arteries
alertness, speech, and communication, preprocedural sedation
may be associated with complications such as dissection or
should be limited to low-dose benzodiazepines to alleviate
spasm (17).
anxiety, if necessary.
Particularly in the early phase of the learning curve, CAS
should be performed in a standardized way using limited and
TECHINQUES familiar equipment (Table 51.3). Lesion length and vessel size
Carotid Angiography to guide equipment choice are usually estimated visually. A
Intubation of the CCA may be achieved with a variety of step-by-step description of the procedure is summarized in
diagnostic catheters. In the presence of a “friendly” aortic Table 51.7. Over a 0.035-in. long steerable wire, a 125-cm 5-Fr
arch, CCA engagement can be achieved with catheter shapes diagnostic catheter is inserted inside an 8-Fr guide catheter or a
such as the Judkins right coronary or the vertebral/Bernstein 90-cm, 6-Fr sheath (Fig. 51.8). The origin of the left CCA or the
catheters. Alternatively, the headhunter or the Benson diagnos- brachiocephalic trunk is intubated with the diagnostic catheter.
tic catheters can be used as routine. For more steep/tortuous Under roadmap, a steerable 0.035-in. guidewire advanced into
aortic arches, shapes such as the Vitek or the Simmons/Side- the distal CCA or, if necessary, into the ECA. The diagnostic
winder or the Benson may be used. The handling of these catheter is advanced over the guidewire into the distal CCA.
catheters is more challenging and may be associated with an Subsequently, the 8-Fr guide catheter or the 6-Fr sheath is
Table 51.7 Carotid Artery Stenting Protocol Using Telescoping Technique, Filter-Based Distal Emboli Protection, and Unfractionated
Heparin as Anticoagulant
l Patient pretreated with aspirin and clopidogrel
l Common femoral arterial access in local anesthesia with a 5-Fr sheath
l Intravenous unfractionated heparin to achieve activated clotting time 250–300 seconds
l 5-Fr pigtail catheter ? aortic arch angiography (LAO 308–408)
l Insert a 125-cm, 5-Fr diagnostic catheter (e.g., JR4) inside a 100-cm, 8-Fr guide catheter (e.g., H1)
l Intubation of the CCA with the diagnostic catheter
l Under roadmap, 0.035-in. guidewire advanced into the distal CCA or ECA, then diagnostic catheter advanced into the distal CCA, then
8-Fr guide catheter advanced over the diagnostic catheter into the mid/distal CCA
l Diagnostic catheter and 0.035-in. guidewire are retrieved
l DSA angiography of the carotid bifurcation and of the intracranial circulation
l ICA lesion passed with filter EPD under roadmap, filter deployed
l 0.5 to 1.0 mg of IV atropine
l Balloon predilatation (if required) with a 3.0–4.0 20 mm, 0.014-in rapid exchange balloon catheter
l Stenting with a nitinol, self-expanding, rapid exchange, 8.0–9.0 30–40 mm stent
l Postdilation (always) with a 5.0–5.5 20 mm, 0.014-in. rapid exchange balloon catheter
l Filter retrieved
l DSA angiography carotid bifurcation and intracranial circulation
l Non-DSA angiography of the CCA during retrieval of guiding catheter/sheath
l Femoral closure device if patient hemodynamically stable
Abbreviations: LAO, left anterior oblique projection; CCA, common carotid artery; DSA, digital subtraction angiography; ECA, external carotid
artery; ICA, internal carotid artery stenosis; EPD, emboli protection device.
[2/7/010/10:21:11] [640–658]
90-cm, 6-Fr sheath, which is advanced into the distal CCA. This
type of engagement is applicable in the presence of a friendly
anatomy of the aortic arch and the CCA and has the advantage
of requiring a smaller femoral access. However, it may not be
used in the presence of severe ECA or distal CCA disease. In
patients with bovine arch, direct intubation of the left CCA
with a guiding catheter (e.g., AL1 8 Fr) may be considered.
Following intubation of the CCA with the guiding cath-
eter or sheath, the lesion, usually located at the origin of the
ICA, is crossed under roadmap with the filter EPD. The device
should be placed in a straight segment of the ICA and filter
apposition to the vessel wall should be verified with angiog-
raphy. On rare occasions, primary crossing of the lesion with
the EPD may not be possible because of lesion complexity. In
these cases, a wire-independent EPD system or a proximal
occlusion type of EPD should be considered. Alternatively, a
buddy wire or predilatation with an undersized balloon (e.g.,
2.0 mm) may be used in order to be able to advance the EPD. To
prevent bradycardia and hypotension associated with balloon
Figure 51.8 Sheath system (90 cm, 6 Fr) for carotid artery stenting angioplasty and stenting, atropine 0.5 to 1.0 mg IV is adminis-
(Cook shuttle sheath, Cook Inc., Bloomington, Indiana, U.S.). tered routinely. Balloon predilatation should be considered
when it is anticipated that advancement of the stent may be
problematic, and in particular, in the presence of high-grade
and/or severely calcified stenosis. Advancement of the stent
advanced over the diagnostic catheter. This step should be within the guiding catheter should be performed slowly, as air
performed under lower magnification in order to have control may be entrapped during the procedure. With respect to stent
of both the guidewire in the distal CCA or ECA and the length, it is recommended to be generous, since the whole
catheter advancement in the aortic arch. diseased segment should be covered and, unlike in the coro-
As an alternative to the telescoping technique, the steer- nary arteries, restenosis is a minor concern in CAS. As a
able 0.035-in. guidewire such as Wholey (Mallinckrodt), Magic consequence, almost invariably the stent will cover the carotid
Torque (Boston Scientific), or Storq (Cordis) can be advanced bifurcation (Fig. 51.9). Even if the ECA seems compromised
under roadmap into the ECA and then the diagnostic 5-Fr following ICA stenting, the patients will invariably remain
catheter can be advanced over the wire into the ECA. Sub- asymptomatic and the ECA stenosis does not need to be
sequently, the steerable guidewire is exchanged for a stiff treated. Postdilatation of the stent is mandatory, but a residual
0.035-in. guidewire such as the Supracore (Abbott) or the stenosis up to 30% is acceptable. Finally, the EPD is retrieved
Amplatz superstiff (Boston Scientific). With the stiff guidewire using the retrieval sheath. Should the flow be compromised
in place in the ECA, the short 5-Fr sheath is exchanged for a during the procedure because of filling of the filter EPD with
Figure 51.9 Carotid artery stenting procedure. Following engagement of the common carotid artery (CCA) with a guiding catheter or long
sheath, the lesion in the internal carotid artery (ICA) is passed with a wire or with the filter emboli protection device (Panel A). Subsequently, a
self-expanding stent is deployed, usually covering the carotid bifurcation (Panels B and C). Thereafter, a balloon postdilatation is performed to
achieve a good stent expansion (Panel D). Abbreviation: ECA, external carotid artery. Source: From Ref. 9.
[2/7/010/10:21:11] [640–658]
debris, then the blood column in the ICA should be aspirated absence of neurologic or angiographic abnormalities, the guide
with an aspiration catheter such as the Export (Medtronic) or catheter/sheath is retrieved into the descending aorta at the
Diver (Invatec) catheters prior to EPD removal to prevent distal time of contrast injection to exclude complications at the level
embolization. of the CCA such as catheter-induced dissections. If the patient
If a distal balloon occlusion EPD is used (PercuSurge is hemodynamically stable, the sheath may be removed and
GuardWire, Medtronic), then the wire containing a central hemostasis may be achieved with a femoral closure device such
lumen that communicates with a low-pressure distal occlusion as Starclose or Perclose (Abbott) or Angioseal (St. Jude Medical,
balloon incorporated into the tip is advanced through the lesion St. Paul, MN, U.S.). Alternatively, the sheath can be pulled few
under roadmap. Following inflation of the balloon distally to hours following the procedure to allow for the effect of heparin
the lesion, the procedure is performed and at the end, prior to to dissipate, and hemostasis can be achieved by manual com-
balloon deflation, an aspiration catheter is used to remove the pression. Antihypertensive medications are discontinued post-
debris suspended in the blood column from the treated vessel. CAS and resumed gradually. Aspirin 100 to 325 mg/day is
If a proximal balloon occlusion system is used (MO.MA or Gore given indefinitely and clopidogrel 75 mg/day is administered
Neuro Protection System), then the device is advanced over a for at least one month.
stiff guidewire previously placed in the ECA. The procedure
can be performed following occlusion of the ECA and CCA
balloon. While in the Gore system there is flow reversal in the Vertebral Stenting
ICA during occlusion time, the MO.MA system requires aspi- Similar to CAS, endovascular treatment of extracranial VA
ration of 60 mL of blood at the end of the procedure prior to stenosis is usually performed under local anesthesia and con-
restoration of blood flow. The handling of occlusive EPD is scious sedation allowing for early detection of neurologic
technically more demanding than the one of filters EPD and symptoms. After arterial access is obtained, unfractionated
should be introduced later in the CAS learning curve. heparin is administered to achieve an activated clotting time
Following retrieval of the EPD, the patient is examined of 250 to 300 seconds. Access is usually obtained at the common
on the catheterization table to detect major neurologic deficits femoral artery. Occasionally, the radial or brachial artery access
and final DSA angiography of the carotid bifurcation and may be preferred based on an unfavorable anatomy at the level
the intracranial vasculature are performed (Fig. 51.10). In the of the aortic arch or in the presence of excessive angulation
between the subclavian and the vertebral arteries. A 6-Fr guide
catheter is advanced over a 0.035-in. wire to obtain a stable
position in the SCA. Usual curves include the vertebral, Judkins
right, and the multipurpose one. If additional stability is
needed, the use of a 7-Fr guide catheter or 7-Fr, 90-cm sheath
may be helpful because it allows the advancement of 0.014- or
0.018-in. buddy wire into the distal SCA. If the long sheath
approach is chosen, then a 0.035-in. exchange length wire is
advanced over a diagnostic catheter positioned in the SCA in
the axillary artery and the long sheath is then advanced over
the wire. Angiographic runs are performed to visualize the
extracranial and intracranial VA and to obtain accurate views
defining the VA lesions and its relation to the SCA.
The use of EPD for VA stenosis is controversial. The
smaller caliber of the VA, the frequently marked angulation
between the VA origin and the SCA, the tortuosity of the
proximal VA segments, and the tendency to develop spasms
are all factors that may cause difficulties in the advancement of
the EPD device. In addition, recovery of the system may also be
difficult in the presence of spasms or unfavorable angulation of
the VA origin, angulation that may be more pronounced fol-
lowing stent deployment. The use of EPD may be considered
for VA with diameters exceeding 3.5 mm, favorable geometric
orientation of the VA origin, and the presence of with ulcerated
target lesions (19).
Under roadmap guidance, the VA lesion is crossed with a
0.014-in. medium-support or extrasupport coronary wire. In
order to obtain stable guiding catheter position, the wire should
be positioned far enough distally in the VA. The tip of the wire
should be visualized during the entire procedure to reduce the
risk of perforation. Stent selection (either self-expanding or
Figure 51.10 Digital subtraction angiography of the carotid bifur- balloon mounted) is based mainly on lesion location. Ostial
cation showing in the left panel a severe stenosis of the internal VA lesion are treated preferably with balloon-expandable cor-
carotid artery (ICA) and in the right panel the result following onary stents because of high radial force, lack of foreshortening,
stenting. Abbreviations: ECA, external carotid artery; CCA, common and low crossing profile. For true ostial lesions, the stent should
carotid artery. protrude 1 to 2 mm into the SCA to allow for optimal lesion
coverage, and a second balloon inflation at high pressure
[2/7/010/10:21:11] [640–658]
The SAPPHIRE study is the only randomized trial com-

paring CEA and CAS performed with the systematic use of
EPD (20). The trial included symptomatic and asymptomatic
patients at high risk for surgery and was designed to prove the
noninferiority of the endovascular approach. The study was
terminated prematurely because of slow enrollment due to
competing CAS registries. Among the 334 patients randomized
(29% of them being symptomatic), major adverse events at one
year occurred in 12.2% in the CAS group and in 20.1% in the
CEA group (P ¼ 0.053). In the actual treatment analysis, the
observed difference reached statistical significance (P ¼ 0.048).
The difference was mainly driven by a reduction in the rate of
myocardial infarction (at 30 days 0.6% in the CAS group vs.
4.3% in the CEA group; P ¼ 0.04). No cranial nerve injury was
observed in the CAS group while this complication occurred in
5.3% of the CEA patients (P < 0.01). The durability of CAS was
documented by a comparable cumulative percentage of major
(1.3% for CAS vs. 3.3% for CEA) and minor (6.1% for CAS vs.
3.0% for CEA) ipsilateral strokes at three years as well as by a
low rate of repeat revascularization during the same period of
time (3.0% for CAS vs. 7.1% for CEA) (25).
The SPACE study sought to prove the noninferiority of
CAS compared with CEA among symptomatic patients. The
use of EPD in the CAS arm was left at the discretion of
the treating physician and was used in 27% of cases. Although
Figure 51.11 Digital subtraction angiography of a proximal steno- the required sample size based on interim analysis was >2400
sis if the left vertebral artery (left panel) treated with a balloon- patients, the trial had to be terminated following the inclusion
expandable coronary stent (right panel). of 1200 patients because of slow enrollment and lack of fund-
ing. The incidence of ipsilateral stroke or death at 30 days was
the primary endpoint of the study and did not differ between
the groups, occurring in 6.8% of cases in the endovascular
group and in 6.3% of patients in the surgical arm (22). At two-
year follow-up, no difference in adverse events between the
should be performed following partial balloon retrieval to two groups could be detected (26).
optimize the apposition of the stent struts at the vessel wall, The EVA-3S was a randomized noninferiority trial com-
the so-called “flaring of the ostium.” Self-expanding stents are paring CAS with CEA in patients with a 60% symptomatic
reserved nonostial lesions in vessels with larger diameters (i.e., carotid artery stenosis. The primary end point was the cumu-
>5.5 mm) (19). In the presence of a severe or calcified lesion, lative incidence of any stroke or death within 30 days after
balloon dilatation prior to stenting may be appropriate. The treatment (23). The protocol did not mandate the use of EPD.
balloon selected should be undersized and shorter than the The performance of CAS without EPD protection in the study
stent placed thereafter (Fig. 51.11). was rapidly halted following the observation that 4/15 patients
treated without protection suffered a stroke, while the propor-
tion of patients treated with protection was 5/58 (OR 3.9; 95% CI,
CLINICAL ASPECTS 0.9–16.7) (27). The entire trial was then stopped prematurely after
Randomized Trials of CEA Vs. CAS the inclusion of 527 patients because of significant increased
Five major—that is, including over 300 patients—randomized event rates in the CAS arm (death or stroke 9.6% in the CAS arm
trials have compared endovascular and surgical carotid revas- and 3.9% in the CEA arm; P ¼ 0.01). At six months, the incidence
cularization. While the SAPPHIRE (Stenting and Angioplasty of any stroke or death was 11.7% in the CAS group and 6.1% in
with Protection in Patients at HIgh Risk for Endarterectomy) the CEA group (P ¼ 0.02). At 4-year follow-up, the death or
trial (20) focused on patients—both symptomatic and asymp- stroke rate still favored CEA, driven by the 30-day events.
tomatic—at high risk for surgery, CAVATAS (CArotid and Beyond 30 days, no difference was observed (28).
Vertebral Artery Transluminal Angioplasty Study) (21), The ICSS randomized 1710 symptomatic patients to CAS
SPACE (Stent-protected Percutaneous Angioplasty of the or CEA (24). The primary end point is the long-term survival
Carotid artery vs. Endarterectomy) (22), EVA-3S (Endarterec- free of disabling stroke. The use of EPD was not mandatory.
tomy vs. Angioplasty in patients with Symptomatic Severe While follow-up is ongoing and expected to be completed in
carotid Stenosis) (23), and ICSS (International Carotid Stenting 2011, the 30-day safety results were recently presented (29). The
Study) (24) enrolled exclusively symptomatic patients. incidence of death, stroke, or periprocedural myocardial
The CAVATAS, performed in the late 1990s, randomized infarction was 8.5% in the CAS group and 5.1% in the CEA
504 symptomatic patients at low to moderate risk for surgery to group (P ¼ 0.004). No difference was observed in the survival
CEA or carotid angioplasty (21). The incidence of death or free of disabling stroke at 120 days.
stroke at 30 days was 10.0% in the endovascular group and A meta-analysis of 10 trials published or presented up to
9.9% in the surgical group. The outcomes between the two May 2009 and reporting 30-day death or stroke has been
groups remained comparable at three years. published. The study included a total of 4648 patients and
[2/7/010/10:21:11] [640–658]
Figure 51.12 Meta-analysis of the randomized trials comparing carotid artery stenting (CAS) with carotid endarterectomy (CEA). Abbre-
viation: OR, odds ratio. Source: From Ref. 9. The references of the smaller trials can be found in the publication.
showed that patients undergoing CAS has a significant increase Large-Scale CAS Registries
in 30-day death or stroke compared to patients treated surgi- The results of seven CAS registries enrolling over 1000 patients
cally (OR 1.60, 95% CI 1.26–2.02) (Fig. 51.12). Statistical tests have been published, for a total of 20,105 patients (Table 51.9).
showed significant heterogeneity in outcomes among the trials All but one were performed in the United States, included
(heterogeneity; P ¼ 0.02) (9). patients at high risk for surgery, and the majority of patients
included were asymptomatic. The good quality of the studies is
Limitations of the CAS vs. CEA demonstrated by the high proportion of mandatory neurologic
Randomized Trials
Current randomized data comparing CAS and CEA have sev-
eral limitations. First of all, the data on asymptomatic patients Table 51.8 Minimal Requirements in Terms of Endovascular
are limited since all but one trial included only symptomatic Expertise in Large-Scale CAS Vs. Endarterectomy Randomized Trials
patients. Second, the use of EPD was mandatory in just one
trial (SAPPHIRE). Third, the minimal endovascular experience CAVATAS (21) Training in neuroradiology and angioplasty
(but not necessarily in the carotid artery)
required per protocol was in most of the trials incredibly low
required. Tutor-assisted procedures
and four of the five large randomized trials (i.e., enrolling over allowed
300 patients) allowed endovascular treatment in the presence SAPPHIRE (20) Procedures submitted to an executive
of a tutor for interventionalists with insufficient experience review committee; CAS periprocedural
(Table 51.8). Other than SAPPHIRE, none of the randomized death or stroke rate had to be <6%
trials would have satisfied the minimum recommended endo- No tutor-assisted procedures allowed
vascular experience according to a multispecialty CAS clinical SPACE (22, 30) At 25 successful CAS or assistance of a
competence statement (31). tutor for interventionalists having
The trials missed the main purpose of randomized testing performed at least 10 CAS
of a new procedure, namely to show that the novel therapy is EVA-3S (23) 12 CAS cases or 5 CAS and 30 cases
of endovascular treatment of supra-
efficacious in the hands of the most skilled operators on
aortic trunks. Tutor-assisted CAS
selected (favorable) patients. In this respect, early testing of allowed for centers not fulfilling minimal
CEA against medical therapy was properly conducted. In the requirements
ACAS trial, for example, patients at high risk for surgery were ICSS (24) A minimum of 50 total stenting procedures,
excluded from the trial and both the centers and the individual of which at least 10 should be in the
surgeons had to demonstrate a 30-day death or stroke rate of carotid artery. Tutor-assisted procedures
<3.0% to be able to enroll. In addition, during the study the allowed for interventionalists with
surgeons were audited in the presence of more than one com- insufficient experience
plication and were allowed to continue enrollment only if no Abbreviation: CAS, carotid artery stenting.
operator-related problem was observed (32,33). Source: From Ref. 9.
[2/7/010/10:21:11] [640–658]
Table 51.9 Thirty-Day Event Rates in Carotid Artery Stenting Registries Enrolling Over 1000 Patients
Industry Surgical Sympt CEC D/S/ D/S D/S
Name Year N sponsored high-risk EPD patients (%) Neurologista adjud. (%) D/S MI (%) sympt (%) asymp (%)
CAPTURE (34) 2007 3500 Yes Yes Mandatory 14 Yes Yes 5.7 6.3 10.6 4.9
CASES PMS (35) 2007 1493 Yes Yes Mandatory 22 Yes Yes 4.5 5.0 NA NA
PRO-CAS (36) 2008 5341 No No 75 55 70 No 3.6b NA 4.3b 2.7b
SAPPHIRE –W (37) 2009 2001 Yes Yes Mandatory 28 Noc Yes 4.0 4.4 NA NA
SVS (38) 2009 1450 No Yes 95 45 No No NA 5.7 NA NA

EXACT (39) 2009 2145 Yes Yes Mandatory 10 Yes Yes 4.1 NA 7.0 3.7
CAPTURE 2 (39) 2009 4175 Yes Yes Mandatory 13 Yes Yes 3.4 NA 6.2 3.0
a
Neurologist, independent pre- and postprocedural assessment by a neurologist.
b
Refers to in-hospital events; EPD ¼ emboli protection devices.
c
Neurologic assessment performed by stroke scale certified staff member.
Abbreviations: sympt, symptomatic; asympt, asymptomatic; CEC adjud., clinical event committee adjudication; D, death; S, stroke; MI, myocardial infarction.
CAROTID AND VERTEBRAL ARTERY INTERVENTIONS
653
[2/7/010/10:21:11] [640–658]
assessment pre- and postprocedure (4/7) and clinical event Data on Vertebral Interventions
committee adjudication of adverse events (5/7). The use of EPD The data on vertebral interventions are virtually limited to
was mandatory in five studies and used in the majority of single-center series, most of them retrospective. Table 51.10
patients in the remaining two. reports the results of the larger series. Although in most series,
The PRO-CAS registry enrolled in German patients with the technical success is greater than 95% and the occurrence of
variable risk for surgery and reported an in-hospital death or periprocedural stroke is a rare event, the placement of a bare
stroke rate of 3.6% (36). Symptomatic and asymptomatic metal stent in the VA is associated with restenosis. The true
patients had an event rate of 4.3% and 2.7%, respectively. In incidence of restenosis is unknown because the follow-up of
the CAPTURE registry, the 30-day stroke/mortality rate was most series was not systematic and the reported results varied
5.7% among 3500 patients, with symptomatic individuals expe- considerably (between 3% and 52%). Although the placement
riencing a stroke rate of 8.9% and asymptomatic patients a rate of drug-eluting stents has proven to be highly efficacious in
of 4.1% (34). The CASES-PMS registry recorded outcomes in restenosis prevention in the coronary artery, the results in the
1493 high-risk patients treated with CAS utilizing EPD reported VA are still sparse. The only two series published with more
a stroke/mortality rate of 4.5%, with a stroke rate of 5.3% in than 25 patients undergoing drug-eluting stent implantation in
symptomatic patients and 3.4% in asymptomatic individuals the VA have reported a promising low restenosis rate of 7%
(35). The SAPPHIRE Worldwide registry reported a 30-day and 12% (51,52).
stroke or death rate of 4.0% in 2001 among high-risk patients, The CAVATAS included also 16 patients with symptom-
with higher event rates in symptomatic compared to asympto- atic VA stenosis that were randomized in equal proportions to
matic patients (adjusted OR 2.4) (37). receive endovascular therapy (balloon angioplasty or stenting)
The SVS registry reported 30-day outcomes among 1450 or best medical treatment alone. An independent neurologist
patients who underwent CAS and 1368 patients treated with followed up the patients for as long as eight years. The trial
surgery (38). In this analysis, the CAS group had significantly failed to show a benefit of endovascular treatment of VA
higher event rates than CEA (death, stroke, or myocardial stenosis, but the number of patients included was small. The
infarction rate 6.4% vs. 2.6%). This analysis was limited by authors of this study concluded that larger randomized trials
the marked imbalances among the groups, the <50% collection are required to determine whether VA stenting is justified in
of 30-day events, the lack of systematic neurologic assessment patients at higher risk of vertebrobasilar stroke (53).
and event adjudication as well as the different definition of
myocardial infarction among the centers.
The results of two large-scale registries enrolling
patients at high risk for surgery, the EXACT (N ¼ 2145) and LIMITATIONS AND COMPLICATIONS
the CAPTURE 2 (N ¼ 4175) studies were recently reported In addition to the limitation of the randomized comparison
(39). The overall 30-day death and stroke rate in the two against CEA just described, the main limitation of CAS is that it
studies were 4.1% and 3.4%, respectively. In the population has not been studied prospectively in specific patient popula-
comparable to AHA guidelines (age <80 years), the pooled tions. Specifically, randomized data compared to CEA in
analysis of the two registries denoted a death or stroke rate asymptomatic patients are lacking. In addition, for surgical
within current recommendations for CEA, namely 5.3% for high-risk patients with asymptomatic carotid disease (e.g.,
symptomatic patients and 2.9% for asymptomatic patients. In SAPPHIRE population), it remains to be demonstrated that
patients 80 years of age, the death and stroke rates in carotid revascularization (both stenting and surgery) is of ben-
symptomatic and asymptomatic patients were 10.5% and efit over best medical management. As previously mentioned,
4.4%, respectively. adequately powered randomized trials in vertebral
Table 51.10 Series of Stenting of the Extracranial Vertebral Artery Including at Least 25 Procedures, Stratified for the Use of BMS or DES
Vessel Type of Technical Periprocedural Follow-up Significant
Series Year treated, N stent successa (%) stroke (mo) restenosisb (%)

Cardiovascular Catheterization and Intervention

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cardiovascular Catheterization and Intervention

Uploaded by

Copyright:

Available Formats

Cardiovascular Catheterization and Intervention Mukherjee

A Textbook of Coronary, Peripheral, and Structural Heart Disease Bates

Cardiovascular Catheterization and Intervention

• Practical suggestions on how to set up a cardiovascular catheterization Marco Roffi, M.D. is

Given the importance of appropriate training and credentialing for clinicians,

This page intentionally left blank

#2010 Informa Healthcare, except as otherwise indicated

No claim to original U.S. Government works

Library of Congress Cataloging-in-Publication Data available on application

For corporate sales please contact: CorporateBooksIHC@informa.com

Typeset by MPS Limited, A Macmillan Company

I would like to thank my wife Nancy and children Andrew, Lyndsay,

To my spouse Muriel and my children Emma, Thomas, and Giulia

This page intentionally left blank

The field of interventional cardiovascular medicine continues to rapidly evolve in

This page intentionally left blank

1. Introduction to cardiac catheterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2. Setting up a catheterization laboratory and equipment considerations . . . . . 13

5. Patient selection, preparation, risks, and informed consent . . . . . . . . . . . . . . . . . . 38

8. Vascular access and closure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

9. Right heart catheterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

12. Hemodynamic assessment for restriction, constriction, hypertrophic

13. Endomyocardial biopsy—indications and procedures . . . . . . . . . . . . . . . . . . . . . . . 136

15. Catheterization of the cardiac venous system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

16. Coronary arterial anatomy: normal, variants, and well-described

19. Cardiac catheterization for pediatric patients ................................ 229

22. Pulmonary angiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270

25. Peripheral arterial angiography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

28. Intravascular ultrasound and virtual histology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355

29. Optical coherence tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370

31. Guiding catheters and wires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404

32. Coronary artery stenting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412

34. Cardiogenic shock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444

36. Saphenous vein grafts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469

38. Complications of PCI: stent loss, coronary perforation

39. Intra-aortic balloon pump counterpulsation, percutaneous

42. Percutaneous therapy for aortic valve disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541

43. Percutaneous therapies for mitral valve disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554

45. Atrial appendage exclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577

48. Embolization for fistulas and AVMs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602

49. Lower and upper extremity intervention ..................................... 610

51. Carotid and vertebral artery interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 640

52. Intracranial stenosis and acute stroke interventions . . . . . . . . . . . . . . . . . . . . . . . . . 659

This page intentionally left blank

J. Dawn Abbott Division of Cardiology, Rhode Island Hospital, Warren Alpert

Alex Abou-Chebl Department of Neurology, University of Louisville, Louisville,

Jorge R. Alegrı́a B. Gill Heart Institute and Division of Cardiovascular Medicine,

Syed Sohail Ali University of Michigan Cardiovascular Center, Ann Arbor,

Sawsan M. Awad Rush University Medical Center, Chicago, Illinois, U.S.A.

Steven R. Bailey Department of Medicine, Division of Cardiology, University of

Colin M. Barker Division of Cardiology, University of Texas Medical School at

Gregory W. Barsness Division of Cardiology, Mayo Clinic, Rochester, Minnesota,

Thomas M. Bashore Division of Cardiology, Duke University Medical Center,

Kenneth L. Baughman{ Division of Cardiovascular Medicine, Brigham and

Jeroen J. Bax Department of Cardiology, Leiden University Medical Center, Leiden,

Yves Laurent Bayard Cardiology, University Hospital Bern, Bern, Switzerland

Matthew C. Becker Department of Cardiovascular Medicine, Cleveland Clinic,

Ryan Berg Division of Cardiology, Saint Louis University School of Medicine,

Lutz Buellesfeld Cardiology, University Hospital Bern, Bern, Switzerland

Adriano Caixeta Interventional Cardiology Department, Hospital Israelita Albert

Qi-Ling Cao Division of Cardiology, Rush University Medical Center, Chicago,

Blase A. Carabello Division of Cardiology, Baylor College of Medicine and the

John D. Carroll Department of Medicine, Division of Cardiology, University of

Joseph P. Carrozza Cardiovascular Medicine–Caritas St. Elizabeth’s Medical